Exogenous metallothionein-IIA promotes accelerated healingafter a burn wound

Natalie M. Morellini, BSc1,2,3; Natalie L. Giles, PhD3,4; Suzanne Rea, MB3,5,6;Katharine F. Adcroft, BSc3,4; Sian Falder, MB3,5,6; Carolyn E. King, PhD1,2; Sarah A. Dunlop, PhD1,2;Lyn D. Beazley, PhD1,2; Adrian K. West, PhD7; Fiona M. Wood, MB3,5,6; Mark W. Fear, PhD3

1. School of Animal Biology, University of Western Australia, Crawley, WA, Australia,

2. Western Australian Institute for Medical Research, Perth, WA, Australia,

3. The McComb Research Foundation, Perth, WA, Australia,

4. Department of Anatomy and Human Biology, University of Western Australia, Crawley, WA, Australia,

5. Royal Perth Hospital, Perth, WA, Australia,

6. Princess Margaret Hospital for Children, Perth, WA, Australia, and

7. NeuroRepair Group, Menzies Research Institute, University of Tasmania, Hobart, Tas., Australia

Reprint requests:Natalie M. Morellini, School of Animal

Biology, University of Western Australia,

Crawley, WA 6009, Australia.

Tel: 161 8 6488 7507;

Fax: 161 8 6488 7527;

Email: moreln01@student.uwa.edu.au

Manuscript received: October 3, 2007

Accepted in final form: May 31, 2008

DOI:10.1111/j.1524-475X.2008.00418.x

ABSTRACT

Severe injury to the epidermal barrier often results in scarring and life-long func-tional deficits, the outcome worsening with a number of factors including timetaken to heal. We have investigated the potential of exogenous metallothioneinIIA (Zn7-MT-IIA), a naturally occurring small cysteine-rich protein, to acceler-ate healing of burn wounds in a mouse model. Endogenous MT-I/II expressionincreased in basal keratinocytes concurrent with reepithelialization after a burninjury, indicating a role for MT-I/II in wound healing. In vitro assays of a humankeratinocyte cell line indicated that, compared with saline controls, exogenousZn7-MT-IIA significantly increased cell viability by up to 30% (p < 0.05), de-creased apoptosis by 13% (p < 0.05) and promoted keratinocyte migration by upto 14% (p < 0.05), all properties that may be desirable to promote rapid woundrepair. Further in vitro assays using immortalized and primary fibroblasts indi-cated that Zn7-MT-IIA did not affect fibroblast motility or contraction(p > 0.05). Topical administration of exogenous Zn7-MT-IIA (2mg/mL) in vivo,immediately postburn accelerated healing, promoted faster reepithelialization(3 days: phosphate-buffered saline (PBS), 8.9� 0.3mm diameter vs. MT-I/II,7.1� 0.7mm; 7 days: PBS 5.8� 0.98mm vs. MT-I/II, 3.6� 1.0mm, p < 0.05)and reduced epidermal thickness (MT-I/II: 45� 4mm vs. PBS: 101� 19 mm,p < 0.05) compared with controls. Our data suggest that exogenous Zn7-MT-IIA may prove a valuable therapeutic for patients with burns and other skininjuries.

Burns are one of the most devastating medical conditions,representing an assault on all aspects of the patient, fromphysical to psychological. Burns affect all ages, in both thedeveloped and developing world, with over 1.25 millionburn injuries annually in the United States.1 Wound heal-ing and subsequent scar formation following a burn injuryis a complex process involving multiple cell types and cell–cell interactions, including inflammation, neovasculariza-tion, collagen deposition, and reepithelialization.2,3 Oneimportant factor relating to the likelihood of hypertrophicscar (HS) formation is time taken to heal: fasterreepithelialization is associated with a reduced severityof scarring.4 Typically, human wounds that heal within14 days have a < 5% chance of developing a HS, whilethose that persist beyond 21 days have a > 80% chance.4

Although many agents such as growth factors and cyto-kines have been tested for their potential to acceleratewound healing after burns, they have met with limited suc-cessful translation into clinical therapeutics.5,6

Here we have investigated a possible role in wound heal-ing for the small, widely distributed cysteine-rich proteins,metallothioneins (MTs). The MT family is best known for

its metal sequestering and antioxidant properties.7 Forexample, they bind zinc and sometimes copper ions, andMT expression is associated with protection against oxi-dative damage, in particular as a result of prolonged UVexposure.8

Of the four isoforms, MT-I and MT-II are often con-sidered together as MT-I/II, because they share significantstructural homology and similarity in temporal and spatialexpression patterns.9 MT-IIA is the major human isoformof this class.10 Several lines of evidence suggest that MT-I/II is involved in wound repair. During healing, MT-I/IIexpression is localized to proliferating epidermal cells,11

while in normal skin it is expressed at low levels in theouter root sheath of hair follicles and the basal cells.12

Topical application of metals such as zinc and copper leadto an up-regulation of MT-I/II expression in wounds andnormal skin, and MT-I/II expression in these cases isclosely associated with proliferative activity.13 The find-ings have led to the suggestion that MT-I/II functions inwound healing by providing increased zinc and othermetal ions for metalloenzymes; this mechanism is pre-sumed to account for the promotion of reepithelialization

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society682

Wound Repair and Regeneration

and cell proliferation associated with MT-I/II expression.However, to date, there has been little investigation of thetherapeutic potential of MT-I/II.

Here we have assessed endogenous MT-I/II expressionand the effects of exogenous Zn7-MT-IIA administeredafter full-thickness cutaneous thermal injury in mice. Wehave also conducted complementary in vitro studies of ker-atinocytes and fibroblasts to establish likely therapeuticconcentrations and investigate possible mechanisms of ac-tion. We report that, as expected, endogenous MT-I/II ex-pression in vivo is increased in keratinocytes duringreepithelialization after a burn injury. In vitro studies indi-cated that exogenous Zn7-MT-IIA application increasedcell viability, decreased apoptosis, and increased migra-tion, while application of Zn alone was able to only par-tially mimic these effects. Furthermore, fibroblastcontraction was unaffected by exogenous Zn7-MT-IIA inin vitro assays. In addition to increases in endogenous Zn-MT-I/II expression during wound healing, application ofexogenous Zn7-MT-IIA accelerated healing in vivo, sug-gesting this may be a useful topical therapeutic or targetfor improving wound healing.

METHODS

MT-I/II immunohistochemistry

Mouse skin sections were obtained at 1, 3, 7, and 11 days(n53 for each time-point) following full-thickness burn in-jury (protocol described below) to examine endogenousMT-I/II expression in keratinocytes during wound healing. En-dogenous MT-I/II expression was also examined in normal(unburned) skin. Sections were stained with a-MT-I/IIantibody (1 : 400; monoclonal mouse, anti-horse metal-lothionenin clone E9; DakoCytomation, Carpinteria, CA),using the Acuity mouse-on-mouse (MoM) kit (Signet Labo-ratories Inc., Dedham,MA) according to the manufacturer’sinstructions. Cells were visualized using DAB/metal complex(Pierce, Rockford, IL).Mean intensity ofMT-I/II expressionwas measured in 10 keratinocytes at each of 10 locationsspanning the burn and surrounding tissue using ImageProPlus (v4.5, Media Cybernetics Inc., Bethesda, MD).

MTs

Zn7-MT-IIA (Bestenbalt, Tallin, Estonia; rabbit-derived,> 98% pure by HPLC) was dissolved in 0.9% sodiumchloride to obtain a final stock concentration of 1mg/mLZn7-MT-IIA. For all in vitro experiments, equivalent con-centrations of zinc sulfate (ZnSO4) were used as zinc con-trols. The zinc control was essential because several lines ofevidence have suggested that zinc is able to exert protectiveeffects independent of MT-I/-II.14,15 For in vivo experi-ments, a final concentration of 2 mg/mL Zn7-MT-IIA wasused (diluted in saline). Final free Zn concentration was160 nM.

In vitro cell lines: Immortalized cells

HaCaT cells, a much studied immortalized human keratin-ocyte cell line,16 NIH/3T3 fibroblasts, an immortalizedmouse fibroblast cell line, were maintained in Dulbecco’s

modified Eagle medium (DMEM)/F12 media supplementedwith 10% fetal bovine serum (FBS, JRH Biosciences,Lenexa, KS), 2mM penicillin (Gibco) and 2mM streptomy-cin (Gibco, Carlsbad, CA). Cell media were renewed twiceweekly with cells being housed in an incubator at 37 1C with5% CO2; cultures were split at 80% confluency.

Primary cells

Primary human fibroblasts were obtained by explant cul-ture from donated skin excised during an elective proce-dure. Fibroblasts between P2 and P6 were used for thecollagen contraction assays.

Cell viability assay

Cell viability assays were performed following treatmentof HaCaT cells with various concentrations of Zn7-MT-IIA. Briefly, cells were seeded at 2.5�103 cells in 100mLvolume per well in a 96-well plate and incubated overnightat 37 1C and 5% CO2. Separate plates were used, for each24, 48, and 72-hour incubation times. Zn7-MT-IIA (exper-imental group) was added at concentrations 0.01, 0.1, 0.3,1, 2, 3, and 10 mg/mL (n57). Equivalent concentrations ofZnSO4 were used as a zinc controls (n57) and saline wasused for the control group (n57). At time-points 0, 24, 48,and 72 hours, 20mL of CellTiter 96sAQueous One solution(Promega, Madison, WI) was added to each well and in-cubated for 3 hours at 37 1C and 5% CO2 before the ab-sorbance was read at 492 nm using a Multiskan Transmitprogram.

Apoptosis protection assay

HaCaT cells were seeded at 2�105 cells in a 2mL volumeper well in six-well plates and incubated overnight at 37 1Cand 5% CO2. Cells were rinsed with phosphate-bufferedsaline (PBS) (pH 7.4) before apoptosis was induced by ex-posing plates to a UV light source for 10 minutes. Cellswere immediately treated with Zn7-MT-IIA at concentra-tions of 0.05, 0.1, 0.5, 1, 2, and 5mg/mL and incubated at37 1C and 5% CO2 for 5 hours. Equivalent concentrationsof ZnSO4 served as zinc controls and untreated cells servedas no treatment controls. At 5 hours following UV expo-sure, cells were removed from each well and labeled withAnnexin-V-FLUOS (Roche, Penzberg, Germany) to iden-tify early apoptotic cells. Cells were analyzed using a fluo-rescent-activated cell sorting Calibur (BD Biosciences,Franklin Lakes, NJ) using 488 nm excitation and 515 nmband-pass filter for fluorescein.

Migration assay

HaCaT cells were seeded at a density of 2�105 cells/mLand NIH/3T3 fibroblasts were seeded at a density of1�105 cells/mL in 12-well plates and were incubated at37 1C with 5% CO2. Once the cell monolayer was conflu-ent, cells were scratched in center of the well with a yellowpipette tip to create a cell-free area in the middle of thewell. Wells were washed with PBS and fresh cell media(DMEM/F12110% FBS) was added. Cells were treatedwith Zn7-MT-IIA at concentrations of 0.01, 0.1, 0.3, 1, 2,3, and 10 mg/mL; equivalent concentrations of ZnSO4 (zinc

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society 683

Metallothionein in burnsMorellini et al.

control); or saline (control). At 6, 12, 24, and 36 hours,scratches were photographed using a light microscope(Olympus 1�51; final objective �4, Olympus, Aizu, Ja-pan). The denuded area was measured using ImageJ soft-ware program.17 Cell migration was calculated bysubtracting the denuded area by the total photographedarea to give cell coverage (mm2). Each treatment groupwas expressed as a percentage of the cell coverage of thesaline (control) group.

Collagen gel contraction assay

Fibroblast-embedded collagen gels were prepared as de-scribed previously.18 Briefly, collagen gel solution was pre-pared by mixing acid soluble ovine collagen (3mg/mL;Ovicolls; CollTech, Perth, Australia), a fivefold concen-trated DMEM (serum free), and a buffer solution (0.05NNaOH, 2.2% NaHCO3, 200mM HEPES) in the ratio7 : 2 : 1, respectively. Cell suspensions of primary humanfibroblasts were dispersed throughout the collagen solutionat a concentration of 1�105 cells/mL. The solution wasgelled at 37 1C for 30 minutes. To prevent the surface fromdrying, 1mL of serum-free DMEM (1�) was added to eachwell and then placed in a humidified chamber at 37 1C, 5%CO2 for 12 hours. Gels were separated from each well,floated in a 6-well plate, and treated with 2mg/mL Zn7-MT-IIA, or equivalent concentrations of ZnSO4 and 0.9% saline(diluted in 1�DMEM) for either a period of 3 hours or con-tinuously (media was not changed). After a 3-hour treat-ment period, gels were floated in DMEMonly. At 12, 24, 36,and 48 hours following treatment, gels were photographedwith Universal Hood II GelDoct XR (Bio-Rad, Hercules,CA). Surface area was measured using Image J softwareprogram.17 The contraction of gels was expressed as apercentage of the initial surface area defined as 100%.

Animals

Adult C57BL/6 mice (Animal Resource Centre, WA, Aus-tralia) were maintained in standard housing with food andwater provided ad libitum. All experiments were approvedby Institutional ethics committees and performed in ac-cordance with the National Health and Medical ResearchCouncil Australian Code of Practice for the Care and Useof Animals for Scientific Purposes.

Procedure

Full-thickness burn wounds were generated following apreviously described protocol.19 Briefly, mice (n540) wereanesthetized in a closed chamber with a continuous flow of2.5–4% isoflurane and given a full-thickness 10mm diam-eter burn by contact with a brass rod (65 g) heated to 95 1Cand applied for 9 seconds. Control animals (n510) sacri-ficed immediately postinjury confirmed histologically thata full-thickness burn was consistently produced. Animalswere administered analgesic (buprenorphine, 0.1mg/kg)subcutaneously immediately postburn. The wounds andtheir surrounds were dressed using a Tegaporet mem-brane dressing (3M Healthcare Ltd., Loughborough, Lei-cestershire, UK) instilled with 200mL PBS (control, n515)or 200 mL PBS containing 2mg/mL Zn7-MT-IIA (experi-mental, n515). Mice were wrapped with a secondary

absorbent dressing with minimal pressure so as not to con-strict breathing and to allow free movement postrecovery.Pain relief was achieved by administering oral analgesia(paracetamol, 0.01mg/mL) in drinking water for 5 daysfollowing the procedure.

Tissue preparation

Mice were euthanized at days 1, 3, 7, 11, and 14 postinjury(n56 at each time-point for experimental [n53] and con-trol groups [n53]). The entire wound, including skin up to3mm peripheral to the wound margins, which had alsobeen exposed to Zn7-MT-IIA or PBS, was removed andfixed in 4% paraformaldehyde. Tissue was processed, par-affin embedded, and serially sectioned at 5mm. For stan-dard histology, one series was stained with hematoxylinand eosin (H&E).

H&E sections were used to measure wound diameter,namely the distance between the wound margins that weredefined by the presence hair follicles, using Stereo Investi-gator 7 software (linear measurement tool, MBFBiosci-ences, Williston, VT) at 3, 7, and 14 days. Epithelialthickness, namely the distance between the basal keratin-ocytes and the outer most epidermal layer, was also mea-sured at the wound center at 14 days.

Statistical analyses

Differences between experimental and control groups wereanalyzed using analysis of variance (ANOVA) and fol-lowed by post hoc analysis (Bonferroni; p < 0.05 as sig-nificant).

RESULTS

MT-I/II immunohistochemistry

In normal mouse skin, endogenousMT-I/II was expressed atlow levels in basal keratinocytes of the epidermal layer (Fig-ure 1A). At 1 day following a full-thickness dorsal skin burn,there was minimal change in MT-I/II expression at thewound margins and far periphery (Figure 1B). At 3 and 7days, however, MT-I/II expression in basal layer of migrat-ing keratinocytes at the wound margins significantly in-creased compared with normal (Figure 1C–D; p < 0.05),with a subsequent up-regulation in the hyperplastic woundcenter at 11 days (after the wound closed; Figure 1E–F).MT-I/II expression in far periphery of the wound did not differfrom normal expression at any time-point (Figure 1A–G).

In vitro

Zn7-MT-IIA promoted HaCaT cell viability at 2, 3, and10 mg/mL (Figure 2A; p < 0.05), an effect not observedwith either saline or ZnSO4 only treatment controls. Fur-thermore, Zn7-MT-IIA was also found to protect HaCaTcells from apoptotic cell death in response to UV irradia-tion (Figure 2B). Zn7-MT-IIA reduced the level of apopto-sis by > 50%, at a concentration of 5 mg/mL (p < 0.05),while ZnSO4 treatment also reduced apoptosis at concen-trations of 1 and 5mg/mL (Figure 2B). In a scratch assay,1, 2, and 3mg/mL Zn7-MT-IIA application promoted

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society684

Metallothionein in burns Morellini et al.

keratinocyte motility, while neither ZnSO4 nor saline con-trols had any effect on the rate of keratinocyte cell migra-tion (Figure 3A–E). In contrast, Zn7-MT-IIA, ZnSO4, orsaline controls did not have any effect on the rate of fibro-blast migration over the injury site (Figure 3F). Further-more, collagen gel contraction assays did not showsignificant differences in the contraction rate between

Zn7-MT-IIA and ZnSO4, nor saline control groups fol-lowing 3 hours or continuous exposure (Figure 4A,B).

In vivo studies

In summary, time to heal was significantly shorter in theexperimental (Zn7-MT-IIA–treated) than in the control(PBS-treated) group. The gross morphology of the woundwas similar in both groups at 1 day but, by 3 days, eschardiameter was smaller in the experimental than in the con-trol group. The difference in wound diameter was moremarked by 11 days (eschar diameter: experimental,1.8� 0.3mm vs. control, 3.4� 0.53mm, p < 0.05). Byday 14, gross morphology did not differ between groups(data not shown).

Wound diameter was also assessed from histologicalsections and showed that wound closure was faster in theexperimental compared with the control group (3 days:experimental, 7.1� 0.7mm diameter vs. control,8.9� 0.3mm; 7 days: experimental, 3.6� 1.0mm diametervs. control 5.8� 0.98mm vs., p < 0.05; Figure 5A–C).Once the wound had become reepithelialized by 14 days,healing was also seen to be improved as assessed by re-duced epidermal thickness (normal: 14� 2mm vs. experi-mental: 45� 4mm vs. control: 101� 19mm, p < 0.05;Figure 5D–G). The observations indicate that the hyper-proliferative response is more advanced following Zn7-MT-IIA treatment with the wound closing and the epider-mis returning to the normal thickness found in noninjuredmouse skin.

DISCUSSION

Rapid repair of the epidermal barrier is essential followingsevere injury in order to minimize scarring and improvefunctional outcomes. In this study, we have shown thatexogenous Zn7-MT-IIA accelerated wound healing in vivo,and propose that it is due to enhanced keratinocyte viabil-ity and migration as well as protection against apoptosis asdemonstrated in vitro. Furthermore, Zn7-MT-IIA did notcause abnormal contraction within fibroblasts.

The increase in endogenous MT-I/II expression in ker-atinocytes migrating across the wound site has also beenobserved in previous skin wound healing studies,11,13 im-plicating a role of MT-I/II in wound repair. Up-regulationof endogenous MT-I/II expression is also observed in thecentral nervous system (CNS) following focal cryole-sions20 and in experimental autoimmune encephalomyeli-tis (EAE), an animal model for multiple sclerosis.21 Theimportance of endogenous MT-I/II is also demonstratedin studies with MT-I/II null mice. Following focal cryoin-jury, MT-I/II null mice showed impaired wound healing20

while following exposure to UVB irradiation, the skin ofMT-I/II null mice showed more edema, postules, sunburncells, and apoptotic cells compared with wild-type mice.8,22

In addition, transgenic mice that overexpress endogenousMT-I (TgMT) are protected against tissue damage and celldeath following brain injury.23 Therefore, endogenousMT-I/II has an important role in wound healing, particu-larly in the CNS, and our study suggests that the observedincrease in epidermal MT-I/II during wound healing may

Figure 1. (A–G) MT-I/II expression in basal keratinocytes at 0,

1, 3, 7, and 11 days following a full-thickness skin burn in

mouse. Mean intensity was measured in 10 keratinocytes at

each of 10 locations spanning the burn and surrounding tissue

using ImagePro Plus. In normal skin, MT-I/II expression is uni-

formly distributed (A). At days 1 (B), 3 (C), and 7 (D) postburn,

keratinocytes are absent from the burned area and MT-I/II ex-

pression was lacking. By day 11, the skin had healed (E). MT-I/II

expression was elevated at the wound margins adjacent to the

burn wound at days 3 and 7 (C, D) and became even more

highly expressed in the wound center at day 11 (E, F). MT-I/II

expression (indicated by arrows) in center of the wound was

significantly greater than the far periphery (F, G). Values are in-

tensity (units)�SEM. np < 0.05. Scale bar: 100mm. CE, wound

center; P, wound periphery; MT, metallothionein.

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society 685

Metallothionein in burnsMorellini et al.

enhance keratinocyte proliferation, migration, and reduceapoptosis, thereby promoting reepithelialization.

Our finding that ZnSO4 only protects against UV-in-duced apoptosis and does not enhance keratinocyte viabil-ity or motility suggests that the MT-I/II protein isimportant not only in delivering zinc for secondary en-zyme activity, but has activity independent of this func-tion. However, we cannot exclude the possibility that Znbound to MT is simply delivered more efficiently to thenecessary enzymes compared with free Zn.

MT-I/II is bound to zinc under natural and physiolog-ical conditions and zinc has been suggested to have tissueprotective effects independent of MT-I/II.15 While studieshave shown that zinc is essential for cell proliferation anddifferentiation in mammalian cells24 and plays a role in in-hibiting apoptosis,25 a recent study has shown that protec-tive effects of zinc against oxidative damage in an SV40fibroblast cell line was mainly due to induction of MT-I/IIrather than direct activity.26 This supports our observa-tions that Zn cannot substitute for the effects of Zn7-MT-IIA on epidermal cells.

Our finding that exogenous Zn7-MT-IIA acceleratedhealing mirrors the improved outcome after adult CNStrauma23,27 and in models of EAE.28 The mode of actionof Zn7-MT-IIA in these models has yet to be established.

Nevertheless, based on the in vitro data presented here andits hypothesized action in the CNS, we suggest that exoge-nous Zn7-MT-IIA acts on basal keratinocytes within hyper-proliferative regions by reducing keratinocyte apoptosisand increasing viability and migration across the woundsite to enhance reepithelialization. Zn7-MT-IIA, however,did not have an effect on fibroblast motility or contraction;thus, the increased rate or wound healing observed in themouse model cannot be attributed to an increase in woundcontraction leading to potential scarring. It is possible thatZn7-MT-IIA may also have an indirect action, for exampleon immune system cells. Studies using labeled exogenousZn7-MT-IIA would help to address this issue.

One possible mechanism of action is that, as for kid-ney29 and retinal cells,30 Zn7-MT-IIA activity in basal ker-atinocytes is mediated via the megalin receptor. However,further studies are required to investigate the role of meg-alin, or other receptors and the related lipoprotein recep-tor-related protein-1.31 A number of reports haveassociated MT-I/II expression with events such as cell pro-liferation and apoptosis,32 with postulated mechanisms in-volving the key MT-I/II properties of zinc chelation/donation, or scavenging of free radicals. The importanceof internalization for exogenous Zn7-MT-IIA activity, thepotential of free radical scavenging and the possibility of

Figure 2. (A–B) Effect of Zn7-MT-IIA admin-

istration in vitro on cell viability (A) and ap-

optosis (B) of HaCaT cells at 72 hours. (A)

Zn7-MT-IIA increased cell viability at 2, 3,

and 10 mg/mL, compared with saline con-

trols, while ZnSO4 did not have an effect.

Values are mean absorbance�SEM. (B)

Zn7-MT-IIA (5mg/mL) and ZnSO4 (1 and

5mg/mL) decreased apoptosis following UV

irradiation exposure. Values are % total

cells, mean�SEM. np < 0.05. MT, metal-

lothionein.

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society686

Metallothionein in burns Morellini et al.

activity through receptor binding will be the subject offuture studies, using labeled and modified Zn7-MT-IIA toidentify the effects of extracellular, as well as intracellular,Zn7-MT-IIA.

It is possible that exogenous Zn7-MT-IIA has other ac-tions in wound healing. It may exert a more generalizedprotective role against oxidative stress and cell damage, asindicated by its up-regulation in the liver during healing of

Figure 3. (A–F) Effect of Zn7-MT-IIA adminis-

tration in vitro on HaCaT and NIH/3T3 cell

migration following a scratch injury. (A) Cell

coverage over a scratch site (A) was measured

24 hours following saline (B), ZnSO4 (C; 2mg/

mL) and Zn7-MT-IIA (D; 2mg/mL) administration

(HaCaT cells). (E) Zn7-MT-IIA increased HaCaT

cell migration at 1, 2, and 3 mg/mL, while ZnSO4

did not have an effect compared with saline

control. (F) In contrast Zn7-MT-IIA or ZnSO4 did

not have any effect on NIH/3T3 fibroblast

migration following scratch injury compared

with saline control. Values are % of saline con-

trol�SEM. np < 0.05. Scale bar5500mm. MT,

metallothionein.

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society 687

Metallothionein in burnsMorellini et al.

severe burns33 and the significant increase in reactive oxy-gen species (ROS) and greater hepatocellular damagecaused in MT-I/II null mice.34 Indeed, treatment with ex-ogenous Zn7-MT in resuscitation fluid following thermalinjury in mice has been shown to reduce oxidative stress.35

Furthermore, Zn-MT was recently used as a topical treat-ment in a rat model of partial burns36 and showed a re-duction in lipid peroxidation and free-radical generation inthe burn wound.

To our knowledge, these studies are the first to assessthe action of exogenous Zn7-MT-IIA on keratinocytes invitro. The enhanced cell migration supports previous stud-ies where MT-I/II promoted chemotactic migration of im-mune cells.37 The increased cell number observed isinteresting in light of in vivo studies reporting Zn7-MT-IIA up-regulation in hyperproliferative epidermal cell

populations in both wound healing and disease.13,38,39

Furthermore, our finding of reduced apoptosis accordswith previous observations in other systems of an antiap-optotic effect of Zn7-MT-IIA, and supports the hypothesisthat Zn7-MT-IIA functions as an antiapoptotic protein.While endogenous MT-I/II has been implicated in prolif-eration and preventing cell death,22,40 we show here thatexogenous Zn7-MT-IIA has the same effects, suggesting apotential therapeutic application.

Overall our results argue strongly that Zn7-MT-IIA willprove a valuable therapeutic to treat burns and other skinwounds. Our in vivo study suggests that Zn7-MT-IIA is sta-ble when administered exogenously to burns. Moreover,we report that the Zn7-MT-IIA concentration used to pro-mote faster healing in vivo was well below the in vitro toxiclevel, suggesting a large therapeutic window. Further

Figure 4. (A–B) Effect of 3 hours and con-

tinuous Zn7-MT-IIA treatment on contrac-

tion of fibroblast-embedded collagen gels.

(A, B) Collagen surface area did not signifi-

cantly differ between Zn7-MT-IIA, ZnSO4 or

saline-treated groups, following 3 hours (A)

and continuous (B) exposure. Values are

percentage of the initial surface area (de-

fined as 100%). MT, metallothionein.

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society688

Metallothionein in burns Morellini et al.

studies are required to determine the optimal dose and timecourse of administration and resolve its mode of action.

Acknowledgments

SAD is an NH&MRC Senior Research Fellow (Grant ID:254670). We thank Michael Archer and Marissa Penrose,School of Animal Biology for assistance with figure prep-aration.

Funding source: The McComb Research Foundation,Perth, WA, Australia and School of Animal Biology Post-graduate project grant, the University of Western Austra-lia, Perth, WA, Australia.

REFERENCES

1. Brigham PA, McLoughlin E. Burn incidence and medicalcare use in the United States: estimates, trends, and datasources. J Burn Care Rehabil 1996; 17: 95–107.

2. Greenhalgh DG. Models of wound healing. J Burn Care Re-habil 2005; 26: 293–305.

3. Martin P. Wound healing—aiming for perfect skin regener-ation. Science 1997; 276: 75–81.

4. Deitch EA. Hypertrophic burn scars: analysis of variables.

J Trauma 1983; 23: 895–8.5. Alemdaroglu C, Degim Z, Celebi N, Zor F, Ozturk S,

Erdogan D. An investigation on burn wound healing in rats

with chitosan gel formulation containing epidermal growth

factor. Burns 2006; 32: 319–27.6. Huang JS, Wang YH, Ling TY, Chuang SS, Johnson FE,

Huang SS. Synthetic TGF-beta antagonist accelerates wound

healing and reduces scarring. Faseb J 2002; 16: 1269–70.7. Palmiter R. The elusive function of metallothioneins. Proc

Natl Acad Sci USA 1998; 95: 8428–30.8. Wang WH, Li LF, Zhang BX, Lu XY. Metallothionein-null

mice exhibit reduced tolerance to ultraviolet B injury in vivo.

Clin Exp Dermatol 2004; 29: 57–61.9. Searle PF, Davison BL, Stuart GW, Wilkie TM, Norstedt G,

Palmiter RD. Regulation, linkage, and sequence of mouse

metallothionein I and II genes.Mol Cell Biol 1984; 4: 1221–30.10. West AK, Stallings R, Hildebrand CE, Chiu R, Karin M,

Richards RI. Human metallothionein genes: structure of the

functional locus at 16q13. Genomics 1990; 8: 513–8.11. Iwata M, Takebayashi T, Ohta H, Alcalde R, Itano Y,

Matsumara T. Zinc accumulation and metallothionein gene

expression in the proliferating epidermis during wound heal-

ing in mouse skin. Histochem Cell Biol 1999; 112: 283–90.

Figure 5. (A–G) Effect of Zn7-MT-IIA on

wound diameter and reepithelialization fol-

lowing a full-thickness burn in vivo. (A, B)

Hematoxylin and eosin (H&E) sections at 7

days following phosphate-buffered saline

(PBS) (A) or Zn7-MT-IIA (B). Wound edges

are located beyond edge of micrograph in

(A) and indicated by arrows in (B). (C)

Wound diameter at 3 and 7 days is reduced

in experimental (Zn7-MT-IIA–treated) ani-

mals compared with controls (PBS;

mean�SEM; np < 0.05). At 14 days,

wounds in both treatment groups had

healed. (D–F) H&E sections in unwounded

skin (D) and at 14 days following burn and

PBS (B) or Zn7-MT-IIA (E) treatment. (F) At

14 days, epidermal thickness (double-

headed arrow; D–F) is increased in

wounded skin compared with unwounded

skin and reduced in experimental (Zn7-

MT-IIA–treated) compared with controls

(PBS, mean�SEM; np < 0.05). Scale

bar5100 mm. MT, metallothionein.

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society 689

Metallothionein in burnsMorellini et al.

12. van der Oord J, Ley M. Distribution of metallothionein innormal and pathological human skin. Arch Dermatol Res1994; 286: 62–8.

13. Lansdown A, Sampson B, Rowe A. Sequential changes intrace metal, metallothionein and calmodulin concentrationsin healing skin wounds. J Anat 1999; 195: 375–86.

14. DiSilvestro RA, Carlson GP. Effects of mild zinc deficiency,plus or minus acute phase response, on CCl4 hepatotoxicity.Free Radic Biol Med 1994; 16: 57–61.

15. Itoh N, Kimura T, Nakanishi H, Muto N, Kobaashi M, Kit-agawa I, Tanaka K. Metallothionein-independent hepato-protection by zinc and sakuraso-saponin. Toxicol Lett 1997;93: 135–40.

16. Boukamp P, Petrussevska RT, Breitkreutz D, Hornung J,Markham A, Fusenig NE. Normal keratinization in a spon-taneously immortalized aneuploid human keratinocyte cellline. J Cell Biol 1988; 106: 761–71.

17. Rasband WS. Image J., U.S. National Institutes ofHealth, Bethesda, MD, USA. http://rsb.info.nih.gov/ij/,1997–2006.

18. Lee YR, Oshita Y, Tsuboi R, Ogawa H. Combination of in-sulin-like growth factor (IGF)-I and IGF-binding protein-1promotes fibroblast-embedded collagen gel contraction. En-docrinology 1996; 137: 5278–83.

19. Bruen KJ, Campbell CA, Schooler WG, deSerres S, CairnsBA, Hultman CS, Meyer AA, Randell SH. Real-time moni-toring of keratin 5 expression during burn re-epithelializat-ion. J Surg Res 2004; 120: 12–20.

20. Penkowa M, Carrasco J, Giralt M, Moos T, Hidalgo J. CNSwound healing is severely depressed in metallothionein I- andII-deficient mice. J Neurosci 1999; 19: 2535–45.

21. Espejo C, Penkowa M, Demestre M, Montalban X, Marti-nez-Caceres E. Time-course expression of CNS inflamma-tory, neurodegenerative tissue repair markers andmetallothioneins during experimental autoimmune encepha-lomyelitis. Neuroscience 2005; 132: 1135–49.

22. Hanada K. Photoprotective role of metallothionein in UV-in-jury—metallothionein-null mouse exhibits reduced toleranceagainst ultraviolet-B. J Dermatol Sci 2000; 23 (Suppl.): S51–6.

23. Giralt M, Penkowa M, Lago N, Molinero A, Hidalgo J.Metallothionein-112 protect the CNS after a focal brain in-jury. Exp Neurol 2002; 173: 114–28.

24. Beyersmann D, Haase H. Functions of zinc in signaling, pro-liferation and differentiation of mammalian cells. Biometals2001; 14: 331–41.

25. Butcher HL, Kennette WA, Collins O, Zalups RK, Koropat-nick J. Metallothionein mediates the level and activity of nu-clear factor kappa B in murine fibroblasts. J Pharmacol ExpTher 2004; 310: 589–98.

26. Santon A, Formigari A, Albergoni V, Irato P. Effect of Zntreatment on wild type and MT-null cell lines in relation toapoptotic and/or necrotic processes and onMT isoform geneexpression. Biochim Biophys Acta 2006; 1763: 305–12.

27. Chung R, Vickers J, Inn ChuahM,West A. Metallothionein-IIA promotes initial neurite elongation and postinjury reac-tive neurite growth and facilitates healing after focal braininjury. J Neurosci 2003; 23: 3336–42.

28. Penkowa M, Hidalgo J. Metallothionein treatment reducesproinflammatory cytokines IL-6 and TNF-alpha and ap-optotic cell death during experimental autoimmune encepha-lomyelitis (EAE). Exp Neurol 2001; 170: 1–14.

29. Klassen RB, Crenshaw K, Kozyraki R, Verroust PJ, Tio L,Atrian S, Allen PL, Hammond TG. Megalin mediates renaluptake of heavy metal metallothionein complexes. Am JPhysiol Renal Physiol 2004; 287: F393–403.

30. Fitzgerald M, Nairn P, Bartlett CA, Chung RS, West AK,Beazley LD. Metallothionein-IIA promotes neuritegrowth via the megalin receptor. Exp Brain Res 2007; 183:171–80.

31. Ambjorn M, Asmussen JW, LindstamM, Gotfryd K, Jacob-sen C, Kiselyov VV, Moestrup SK, Penkowa M, Bock E,Berezin V. Metallothionein and a peptide modeled aftermetallothionein, EmtinB, induce neuronal differentiationand survival through binding to receptors of the low-densitylipoprotein receptor family. J Neurochem 2008; 104:21–37.

32. Shimoda R, Achanzar WE, Qu W, Nagamine T, Takagi H,Mori M, Mori M, Waalkes MP. Metallothionein is a poten-tial negative regulator of apoptosis. Toxicol Sci 2003; 73:294–300.

33. Ding HQ, Zhou BJ, Liu L, Cheng S. Oxidative stress andmetallothionein expression in the liver of rats with severethermal injury. Burns 2002; 28: 215–21.

34. Cho K, Adamson L, Jeong J, VanHook T, Rucker R, Green-halgh D. Alterations in the levels of metallothionein and met-als in the liver, and unique serum liver enzyme response inmetallothionein knock-out mice after burn injury. Pathobiol-ogy 2004; 71: 223–30.

35. Zhou Z, Ding H, Qin F, Liu L, Cheng S. Effect of Zn7-met-allothionein on oxidative stress in liver of rats with severethermal injury. Acta Pharmacol Sin 2003; 24: 764–70.

36. Yu D, Qin F, Sun Y. Effects of metallothionein (MT) onburned skin of rats. Chung-Hua Cheng Hsing Shao ShangWai Ko Tsa Chih 1999; 15: 92–4.

37. Yin X, Knecht DA, Lynes MA. Metallothionein mediatesleukocyte chemotaxis. BMC Immunol 2005; 6: 21.

38. Lim D, Phan TT, Yip GW, Bay BH. Up-regulation of met-allothionein isoforms in keloid keratinocytes. Int J Mol Med2006; 17: 385–9.

39. Rossen K, Haerslev T, Hou-Jensen K, Krag Jacobsen G.Metallothionein expression in basaloid proliferations overly-ing dermatofibromas and in basal cell carcinomas. Br JDermatol 1997; 136: 30–4.

40. Hanada K, Sawamura D, Hashimoto I, Kida J, NaganumaA. Epidermal proliferation of the skin in metallothionein-null mice. J Invest Dermatol 1998; 110: 259–62.

Wound Rep Reg (2008) 16 682–690 c� 2008 by the Wound Healing Society690

Metallothionein in burns Morellini et al.