www.fems-microbiology.org

FEMS Microbiology Letters 240 (2004) 105–110

Investigation of morphological changes to Staphylococcus aureusinduced by ovine-derived antimicrobial peptides using

TEM and AFM

Rachel Claire Anderson, Richard G. Haverkamp, Pak-Lam Yu *

Institute of Technology and Engineering, Massey University, Private Bag 11-222, Palmerston North, New Zealand

Received 18 July 2004; received in revised form 9 September 2004; accepted 15 September 2004

First published online 2 October 2004

Edited by A.M. George

Abstract

The effect of two ovine-derived peptides on the morphology of Staphylococcus aureus NCTC 4163 cells were investigated using

transmission electron microscopy and atomic force microscopy. Both techniques showed that SMAP29, an a-helical peptide,

induced cell lysis, whereas OaBac5mini, a proline/arginine-rich peptide, caused no observable morphological changes. This is con-

sistent with previous experimental work which indicated that SMAP29 caused cell death and induced cell lysis, whereas OaBac5mini

acted by interacting with the inner cellular contents.

� 2004 Federation of European Microbiological Societies. Published by Elsevier B.V. All rights reserved.

Keywords: Antimicrobial peptides; Cathelicidins; Transmission electron microscopy; Atomic force microscopy; Ovine blood

1. Introduction

Antimicrobial peptides are part of the innate immune

system of all living things. These peptides act as the first

line of defence against invading micro-organisms. This

work focuses on ovine antimicrobial peptides becauseovine blood is readily available in New Zealand and

has the potential to be separated into a number of

high-value products, including the antimicrobial pep-

tides isolated from the blood neutrophils [1].

For this work two synthesised ovine-derived peptides

were used (Table 1). SMAP29 is a very potent a-helicalpeptide [2] that was predicted from ovine cDNA [3,4],

but has not yet been isolated from ovine cells. OaBac5-mini is a truncated version of OaBac5, comprising the

0378-1097/$22.00 � 2004 Federation of European Microbiological Societies

doi:10.1016/j.femsle.2004.09.027

* Corresponding author. Tel.: +64 6 350 5027; fax: +64 6 350 5604.

E-mail address: p.yu@massey.ac.nz (P.-L. Yu).

first 24 N-terminal residues. OaBac5 is a 51-residue, pro-

line and arginine-rich peptide that was inferred based on

a cDNA sequence [5]. Since then, three variants of Oa-

Bac5, as well as the originally predicted molecule, have

been isolated from sheep neutrophils [6,7]. OaBac5 is

made up of a 6-residue N-terminus, followed by twocopies of a 16-residue repeat and a 5-residue C-terminus

[7]. OaBac5mini, the truncated version of OaBac5, is

made up of the six N-terminal residues, one copy of

the 16-residue repeat, and the first two residues of the

second repeat. The full-length molecule could not be

synthesised, so this truncation was chosen because it

has been shown that the bovine peptide Bac7 retained

its activity when truncated similarly [8]. It has previouslybeen shown that OaBac5mini displays similar antimicro-

bial activity to the parent molecule, OaBac5, so it is an

appropriate model to use to represent the naturally

occurring peptide [9].

. Published by Elsevier B.V. All rights reserved.

Table 1

Sequences of ovine-derived antimicrobial peptides used for this

research

Peptide Sequence

SMAP29 RGLRRLGRKIAHGVKKYGPTVLRIIRIA–NH2

OaBac5mini RFRPPIRRPPIRPPFRPPFRPPVR–NH2

106 R.C. Anderson et al. / FEMS Microbiology Letters 240 (2004) 105–110

Previous worked showed that SMAP29 and OaBac5-

mini had different mechanisms-of-action [9]. The aim of

this research was to investigate whether the test peptides

induced changes in the bacterial cell morphology to help

further understand the mechanisms used by the pep-

tides. The previous results indicated that SMAP29

caused rapid cell death by depolarising the cytoplasmic

membrane, which resulted in leakage of the cellular con-tents, and some complete cell lysis. In contrast, OaBac5-

mini appeared to inhibit cell division by interacting with

the inner cellular contents. Therefore, it was hypothe-

sised that SMAP29 would cause the morphology of

the cells to change significantly, whereas OaBac5mini

would cause little or no change.

To get a good understanding of the morphology of

the peptide treated cells two techniques, transmissionelectron microscopy (TEM) and atomic force micros-

copy (AFM), were used. TEM gives an electron density

image of cell cross sections, whereas AFM gives topo-

logical images of the cell surface. Together these tech-

niques should give a good overall model of the cell

morphology.

2. Materials and methods

2.1. Peptide synthesis

The two ovine-derived peptides were synthesised, by

N-(9-fluorenyl)methoxy carbonyl solid-phase peptide

synthesis using a model 432A synthesiser (Applied Bio-

systems Inc, Foster City, CA), at the University of

British Columbia Nucleic Acid/Protein Service facility.

The purities of the peptides were confirmed to be at

least 99% using high-performance liquid chromatogra-

phy and mass spectroscopy analysis. SMAP29 was syn-thesised as a 28-amino acid amidated peptide because

this is thought to be its natural form [2]. The full

OaBac5 molecule could not be synthesised, due to

the large number of proline residues, so a shortened

version (the first 24 amino acids) called OaBac5mini

was created. Both peptides were amidated at their

C-termini.

2.2. Transmission electron microscopy

A log phase culture of Staphylococcus aureus NCTC

4163 in Mueller–Hinton broth was split into 1.5 mL

aliquots. The cells were collected by centrifugation

(10000 rpm, 5 min) and resuspended in peptone water.

Six samples were prepared; two untreated, two treated

with SMAP29 (4 lg/mL) and two treated with OaBac5-

mini (64 lg/mL). The samples were incubated at 37 �Cfor 1 h. The cells were collected by centrifugation(10000 rpm, 5 min) to remove the peptone water. To

fix the cells, 2% glutaraldehyde in 0.1 M cacodylate

was added and the samples were incubated at 4 �Cfor 1 h. The cells were collected by centrifugation

(10000 rpm, 5 min) and washed twice with 0.1 M phos-

phate buffer. To postfix the cells, 1% osmium tetraox-

ide was added and the samples were left at room

temperature for 1 h. The samples were dehydrated withgraded ethanol solutions (30% ethanol for 10 min, 60%

ethanol for 10 min, 90% ethanol for 10 min, 100% eth-

anol for 10 min, 100% ethanol for 1 h), embedded in

Procure 812 resin and left to polymerise (over the

weekend). From each sample 10 thin slices (approxi-

mately 100 nm) were cut with a diamond knife and

stained with uranyl acetate and lead citrate on grids.

Each of these sections was examined with a PhilipsEM201 80 kV Transmission Electron Microscope and

images were taken with a 35-mm camera.

2.3. Atomic force microscopy

A log phase culture of S. aureus NCTC 4163 in

Mueller–Hinton broth was split into 1 mL aliquots.

The cells were collected by centrifugation (10000 rpm,5 min) and resuspended in peptone water. Three sam-

ples were prepared; one was untreated, one was treated

with SMAP29 (4 lg/mL for both organisms) and one

was treated with OaBac5mini (64 lg/mL). The samples

were incubated at 37 �C for 30 min. The cells were col-

lected by centrifugation (10000 rpm, 5 min) to remove

the peptone water. A number of different methods were

tried to secure the cells for AFM imaging. These in-cluded passing the samples through polycarbonate

membranes (SPI-Pore Filter, 13 mm diameter, 0.8 lmpores) to entrap the cells in the pores and immobilising

the cells in agarose gel (5% agarose). The most success-

ful method involved drying the culture on a glass slide.

The treated samples were washed with and resuspended

in water, and then diluted 100-fold. About 10 lL of the

sample was dropped onto a glass microscope slide andallowed to dry at room temperature. For each sample

duplicate slides were prepared. The dried samples typi-

cally had a diameter of 5–8 mm. To image the samples

an Asylum Research MFP-3D scanning probe micro-

scope was used. The tips used were either NT-MDT

CSG01 or Olympus TR400PSA which typically have

a spring constant in the range 0.01–0.08 N/m. The areas

for imaging were chosen randomly from the totalsample area.

R.C. Anderson et al. / FEMS Microbiology Letters 240 (2004) 105–110 107

3. Results and discussion

3.1. Transmission electron microscopy

S. aureus 4163 NCTC cells were treated with

SMAP29 or OaBac5mini at concentrations four timestheir minimal inhibitory concentrations (MICs) [9], be-

fore they were fixed, cut into sections and examined

using TEM. For each treatment duplicate samples were

prepared and from each sample 10 sections were cut

and processed. The electron density images given in

Fig. 1 shows the typical results. The dark areas show

where the sample had a high electron density and the

light areas show where the sample had a low electrondensity.

Fig. 1. Transmission electron microscope images taken of S. aureus 4163 N

arrows point to cell walls that have separated from the cytoplasm. The green

interpretation of the references to colour in this figure legend, the reader is

The images of the samples treated with SMAP29 were

greatly different to those of the untreated cells. The un-

treated cells were uniformly shaped, with intact cell

walls; whereas many of the SMAP29 treated cells had

their cell walls missing. In Fig. 1 the green arrows point

to cells without cell walls, called ghost cells, and the yel-low arrows point to cell walls that have separated from

the cells. The images show that in many cases the cyto-

plasmic contents were leaking out of the cells. Previous

experiments showed that SMAP29 causes cell death

and induces lysis of Gram-negative cells [9]. These re-

sults indicate that SMAP29 acts on Gram-positive cells

in a similar way to how it acts on Gram-negative cells.

In contrast, the images of the samples treated withOaBac5mini show fewer damages than those treated

CTC cells treated with SMAP29 and OaBac5mini for 1 h. The yellow

arrows point to ghost cells that are not surrounded by a cell wall. (For

referred to the web version of this article.)

108 R.C. Anderson et al. / FEMS Microbiology Letters 240 (2004) 105–110

with SMAP29 (Fig. 1). Unlike the SMAP29 treated

sample, there were no ghost cells without cell walls, or

cells that appeared to be leaking their cellular contents.

However, some cell wall debris was present (yellow ar-

rows). This is consistent with the idea that OaBac5mini

does not cause cell death and induce lysis [9]. However,the cells treated with OaBac5mini appeared to be swol-

len and the samples had a high proportion of cells con-

taining septa, compared with the control samples. This

may be due to OaBac5mini stopping the cell division

process at this point, whereas the control cells moved

on from this state and continued to divide as normal.

Fig. 2. Far away AFM images of S. aureus NCTC 4163 cells on a glass slide

left are the height images and those on the right are the deflection images.

This is consistent with previous work that indicated that

the mechanism-of-action of OaBac5mini involved the

inhibition of cell division [9].

3.2. Atomic force microscopy

S. aureus NCTC 4163 cells were treated with

SMAP29 or OaBac5mini at four times their minimal

inhibitory concentrations (MICs) [9], before they were

dried at room temperature on glass slides and imaged

using AFM. For each sample duplicate slides were made

and numerous randomly chosen parts of each slide were

treated with SMAP29 and OaBac5mini for 30 min. The images on the

R.C. Anderson et al. / FEMS Microbiology Letters 240 (2004) 105–110 109

imaged. The images shown are typical. The results are

shown as both an overview of numerous cells in Fig. 2

(30 · 30 lm) and as a close up of one or two cells in

Fig. 3 (2 · 2 lm). The images on the left are the height

images. In these images the light areas are parts of the

sample that were high and the dark areas are parts ofthe sample that were low. The images on the right are

the deflection images. In these images the light areas

are parts of the sample where the height was increasing

and the dark areas are parts of the sample where the

height was decreasing as it was scanned by the AFM

tip from left to right. The apparent sloping left hand side

Fig. 3. Close up AFM images of S. aureusNCTC 4163 cells on a glass slide tr

are the height images and those on the right are the deflection images.

of the cells is an artefact of the method due to contact

with the side, rather than point, of the tip.

The AFM images of the untreated cells showed that

the slide was covered with a thin layer of material. This

was probably the remains of the growth media. Directly

around the bacterial cells there were zones where thismaterial was not present. This can be seen clearly in

Fig. 2. The material on the edge of the clear zones was

different to that which covered the rest of the slide. It

was in clumps that were higher than the thin film cover-

ing. These rings of built-up material around the cells can

be seen in Fig. 3. It appears that the cells repelled the

eated with SMAP29 and OaBac5mini for 30 min. The images on the left

110 R.C. Anderson et al. / FEMS Microbiology Letters 240 (2004) 105–110

media solids during the drying process, but the reason

for this is unknown.

The AFM images of the cells treated with OaBac5mi-

ni were similar to those of the untreated cells. The clear-

ings directly around the cells (Fig. 2) and the clumpy

material in rings were both present (Fig. 3). This is con-sistent with the idea that the mechanism-of-action of

OaBac5mini does not induce the bacterial cells to lyse

[9].

In contrast, the AFM images of the cells treated with

SMAP29 were different to those of the cells that were

untreated or treated with OaBac5mini. There were a

lot fewer whole cells present on the slide (Fig. 2). In

the case of the untreated and OaBac5mini treated sam-ples, each of the areas of the sample chosen randomly

for imaging contained numerous cells, whereas for the

SMAP29 treated samples, the randomly chosen areas

for imaging contained fewer, and in some cases no cells.

This indicated that the majority of the cells had been

lysed. This high proportion of lysed cells may be the re-

sult of the drying process. SMAP29 may have weakened

the cell walls and caused the cells to be leaky, whichmade them prone to lysis during drying.

Unlike the other samples, which only had built up

material in rings around the cells, the SMAP29 treated

sample contained a number of piles of material. This

material was probably cell debris from lysis induced

by the peptide. The cells themselves did not appear to

be morphologically different to those of the untreated

sample. Interestingly, around the cells that were intactthere were no clearings like those in the other samples.

The reason for this is unknown.

As predicted there were significant differences in the

culture treated with SMAP29 to that treated with Oa-

Bac5mini. This gives further evidence to support the re-

sults in the previous work [9] which showed that the two

ovine-derived peptides had different mechanisms of

action.

Acknowledgements

This work was supported by MeatNZ and the Mas-

sey University Research Fund (MURF). R.C.A. was a

recipient of the MeatNZ Doctoral Scholarship. The syn-

thesised peptides were kindly supplied by the R.E.W.Hancock Laboratory (Department of Microbiology

and Immunology, University of British Columbia, Van-

couver, Canada). We thank HortResearch (Palmerston

North, New Zealand) for the use of the TEM and Aaron

Hicks (Institute of Veterinary, Animal and Biomedical

Sciences) for helping with the sample preparation.

References

[1] Anderson, R.C. and Yu, P.L. (2004) Ovine antimicrobial peptides:

new products from an age-old industry. Aust. J. Agr. Res. 15, 69–75.

[2] Skerlavaj, B., Benincasa, M., Risso, A., Zanetti, M. and Gennaro,

R. (1999) SMAP-29: a potent antibacterial and antifungal peptide

from sheep leukocytes. FEBS Lett. 463, 58–62.

[3] Bagella, L., Scocchi, M. and Zanetti, M. (1995) cDNA sequences

of three sheep myeloid cathelicidins. FEBS Lett. 376, 225–228.

[4] Mahoney, M.M., Lee, A.Y., Brezinski-Caliguri, D.J. and Huttner,

K.M. (1995) Molecular analysis of the sheep cathelin family reveals

a novel antimicrobial peptide. FEBS Lett. 337, 519–522.

[5] Huttner, K.M., Lambeth, M.R., Burkin, H.R., Burkin, D.J. and

Broad, T.E. (1998) Localisation and genomic organisation of sheep

antimicrobial peptide genes. Gene 206, 85–91.

[6] Shamova, O.V., Brogden, K.A., Zhao, C., Nguyen, T., Kok-

ryakov, V.N. and Lehrer, R.I. (1999) Purification and properties of

proline-rich antimicrobial peptides from sheep and goat leuko-

cytes. Infect. Immun. 67, 4106–4111.

[7] Anderson, R.C. and Yu, P.L. (2003) Isolation and characterisation

of proline/arginine-rich cathelicidin peptides from ovine neutroph-

ils. Biochem. Biophys. Res. Commun. 312, 1139–1146.

[8] Sadler, K., Eom, K.D., Yang, J.L., Dimitrova, Y. and Tam, J.P.

(2002) Translocating proline-rich peptides from the antimicrobial

peptide bactenecin 7. Biochemistry 41, 14150–14157.

[9] Anderson, R.C., Hancock, R.E.W. and Yu, P.L. (2004) Antimi-

crobial activity and bacterial membrane interaction of ovine-

derived cathelicidins. Antimicrob. Agents Chemother. 48, 673–676.