Journal of Internal Medicine 2003; 254: 197–215

REVIEW

Antibacterial peptides: basic facts and emerging concepts

H. G. BOMAN
From the Microbiology and Tumor Biology Center, Karolinska Institutet, Stockholm, Sweden

Abstract. Boman HG (Microbiology and Tumor allergy. Several antibacterial peptides are being
Biology Center, Karolinska Institutet, Stockholm, developed as drugs.
Sweden). Antibacterial peptides: basic facts and
Keywords: cathelicidins, cecropins, defensins, endo-
emerging concepts (Review). J Intern Med 2003;
genous antibiotics, innate immunity, LL-37, morbus
254: 197–215.
Kostmann, peptide-based defence.
Antibacterial peptides are the effector molecules of
innate immunity. Generally they contain 15–45 Abbreviations: Drosophila, Drosophila melanogaster,
amino acid residues and the net charge is positive. the fruit fly used by geneticists; Cecropia, the Cecropia
The cecropin type of linear peptides without cys- moth, Hyalophora cecropia; NF-jB, nuclear factor jB,
teine were found first in insects, whilst the defensin one of many transcription factors; IjB, inhibitor
type with three disulphide bridges were found in of NF-jB, destroyed after phosphorylation; jB,
rabbit granulocytes. Now a database stores more upstream DNA motif that binds the activated NF-jB
than 800 sequences of antibacterial peptides and complex; Toll, Drosophila gene, coding for a receptor,
proteins from the animal and plant kingdoms. first found in development; TLR, toll-like receptors,
Generally, each species has 15–40 peptides made mammalian proteins resembling toll; Imd, Drosophila
from genes, which code for only one precursor. The gene needed for induction of antibacterial peptides;
dominating targets are bacterial membranes and HND, human neutrophil defensins (a-defensins),
the killing reaction must be faster than the growth sometimes only ND; HBD, human b-defesins, pro-
rate of the bacteria. Some antibacterial peptides are duced by epithelia,, sometimes only HD; Cathelicidin,
clearly multifunctional and an attempt to predict group name for mammalian antibacterial peptides
this property from the hydrophobicity of all amino with the cathelin propart; LL-37, the linear
acid side chains are given. Gene structures and C-terminal effector of cathelin-LL-37 (hCAP-18);
biosynthesis are known both in the fruit fly Dro- CRAMP, the mice homologue of LL-37, a cathelicidin;
sophila and several mammals. Humans need two PGRP, peptidoglycan recognition protein, found in
classes of defensins and the cathelicidin-derived most animals; LPS, bacterial lipopolysaccharide;
linear peptide LL-37. Clinical cases show that defi- TNF-a, tumour necrosis factor alpha; VIP, vasoactive
ciencies in these peptides give severe symptoms. intestinal peptide, a neurotransmitter affecting
Examples given are morbus Kostmann and atopic kidney function.

review is different: it intends to explain why most

To be ‘born with’
of us stay healthy. The explanation is the protec-
Most reviews in clinical journals deal with diseases tion offered by the innate immune system; we
and what goes wrong in the human body. This share such machinery with all other animals and
Ó 2003 Blackwell Publishing Ltd 197
198 H. G. BOMAN

also plants. The main components of innate (a)

immunity – something we all are ‘born with’ –
100

Log viable of Bo-3N (% CFU)

are a set of peptides with antibacterial activity. Frog blood and serum
These peptides – some with family names like
cecropins, defensins or cathelicidins – are potent 0.1 mg mL–1
10
antibiotics containing some 15–45 amino acid
residues and as a rule have a positive net charge.
0.2 mg mL–1
Skin
When I say ‘born with’ I mean it literally because secretion
the synthesis of enteric defensins in foetal tissue 1

starts 13.5–17 weeks after gestation [1]. More-

over, a very recent report documents that LL-37 0.3 mg mL–1
0
(a cathelicidin) and the defensin families are 0 5 10 15 20 25 30 35
present both in Vernix Caseosa and amniotic fluid Incubation time at 30 ˚c (min)
[2]. As it turns out, the two classes of defensins and
LL-37 are the components most essential also for (b) 4 Dose: 20 million cells at zero time
protecting our adult life from bacterial infections.
As far as is known, an inherited loss of LL-37 has Cortison-pretreated frog

Log viable Bo-3N (CFU)

3
only been survived by four people [3]. I will return
to this finding later.
2
What matters is the killing rate El. stimulated frog

Why did you not hear about that before? One 1

answer is that ‘innate’ immunity may not be a Control (wild frog)
very well chosen name – it would have been more
0
adequate to call it ‘instant’ immunity, because fast 0 10 20 30 40 50 60
killing of bacteria is a hallmark of peptide-based Time after infection (min)
defence. Bacteria in the right environment can
Fig. 1 Rapid elimination of frog bacteria in vitro and in vivo. A
double in number every 20 min. For a surviving
number of frog bacteria were isolated from the skin and mouth of
host it is obvious that an effective defence must be several wild frogs, species Rana esculenta and Bombina orientalis.
faster than the growth rate of an invader. Thus, All animals tested carried Aeromonas hydrophila and strain desig-
my first illustration (Fig. 1) shows parallel in vitro nations were made to indicate the frog species of the isolate, in
this case B. orientalis. (a) Different concentrations of skin secretion
and in vivo experiments in which 20 million bac-
from the frog R. esculenta were used to study the rate of killing of a
teria were eliminated from the mouth of a frog. The naladixic acid resistant mutant A. hydrophilia (Bo-3N) at 30 °C.
strain of bacterium used (Aeromonas hydrophila) The number of live bacteria at different times is given in log scale.
was isolated as a species normally carried by the Frog blood and serum were controls. (b) Mouth infection of three
R. esculenta with A. hydrophila. One frog was untreated, one
frog and the control with a cortisone-pretreated
el-stimulated before the infection and one pretreated with a
frog shows that the frog did not swallow the glucocorticoid cream on the skin. The strain used was Bo-3N and
bacteria [4]. It also shows that cortisone blocks the bacterial titre in the mouth was followed by pipetting 20 lL of
de novo synthesis of antibacterial peptides in the LB medium in the mouth and quickly withdrawing 5 lL which
was plated in 3 mL soft agar on an LB plate with nalidixic acid
frog. Our natural flora of bacteria is another reason
(20 lg mL)1). The results are given in log scale as per cent of the
why we need a very fast defence against bacteria. maximum value. Cortison is known to block de novo synthesis of
We carry nearly 2 kg microbes (mainly bacteria) in all antibacterial peptides in the frog (blocking of NF-jB by
the digestive system and some 200 g on outer induction of IjBa synthesis).
surfaces. The control of this flora is dependent both
on the turnover of antibacterial peptides in the host
and on a dynamic equilibrium in the flora itself – host. It is the microbial circulation of carbon and
an ecosystem with a certain internal control. When nitrogen: ‘For dust thou art and unto dust shalt
death occurs these two mechanisms stop and the thou returne’ (God to Adam after the Fall, Genesis
bacteria begin to multiply and brake down the dead 3:19).
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
 REVIEW: ANTIBACTERIAL PEPTIDES 199

accord with this new perspective being accepted,

A time when nobody cared
the number of peptide sequences began to increase.
How did ‘instant’ immunity start? We reported in In January 1994 some 50 sequences were known.
1972 that adult Drosophila flies could be vaccinated Today a database in Trieste contains more than 800
by an injection of harmless bacteria which protected sequences for antimicrobial peptides and proteins
the flies against an otherwise lethal dose of Pseudo- from animals and plants [13]. Of these, 346 are
monas aeruginosa [5]. The experiments indicated that linear peptides of which 278 can form a-helices.
the flies had a cell-free defence, but it was difficult to There are 330 peptides with S–S bridges and
follow-up. We had to switch from Drosophila to these include a-defensins number 54 and the
pupae of a large insect. The Cecropia moth was b-defensins 68.
ideal; only one generation per year and long winter There are many specialized reviews on immunity
hibernation (diapause) in the pupal stage. To use in insects [14] or other species such as frogs [15]
dormant pupae was important, because during and mice [16]. Many reviews are related to defensins
dormancy the general metabolism of the insect is [17], epithelial host defence [18] or intestinal
reduced to a few percent, which means that the immunity [19, 20], and a few have broader aims
synthesis of most proteins is shut off. Just like the [21–23]. Insect and mammalian immunity are both
flies, ‘sleeping’ pupae (stored in a refrigerator) could linked to malaria and the complex life cycle of the
turn on their immune system after a vaccination. parasites is discussed at the end of this article.
This gave us a significant biological enrichment of
immune protein in the haemolymph (the insect
Wake-up time
blood, nearly 1 mL per pupae).
After 8 years we obtained two heat-stable peptides Antibacterial peptides can be grouped into five
that killed and lysed bacteria very fast [6]. The classes with different 3-D structures [21, 24]. An
killing reaction was insensitive to pH or salts, additional class may be added for the h-defensins,
properties that made an enzymatic mechanism of which are circular peptides made from two genes
action less likely. Sequences followed a year later with a novel biosynthesis [25]. However, so far the
and showed two linear basic peptides with 35 and scientific interest has been focused mainly on three
37 residues and no cysteine [7]. Cecropia has also of the classes: (i) linear peptides free of cysteines
an additional cecropin, one lysozyme and two larger and often with an a-helical and amphipathic
antibacterial proteins, the attacins, all purified in the solution structure, (ii) peptides with three disul-
beginning of the 1980s. phide bonds giving peptides with a flat dimeric
In 1983, Lehrer and Selsted purified two antibac- b-sheet structure and (iii) peptides with unusual
terial peptides with three intralinked cysteines from bias in certain amino acids, such as proline,
rabbit lung macrophages [8]. These were later arginine, tryptophan or histidine (see Fig. 2 for
named defensins [9]. selected sequences of each type).
When something unexpected but universal is Some antibacterial peptides show both antifungal
found, fellow scientist most often say, ‘Oh, it is just and antiviral activity and, for this reason, the
something special with that insect or that cell’. peptides are often referred to as ‘antimicrobial’. I
Thus, those involved had to continue with more have kept the term antibacterial because I believe
peptides isolated from other insects and from that this function has been the driving force during
granulocytes of other mammals. The interest in this evolution. Fungi usually do not cause wild mam-
field increased after Zasloff’s isolation of magainins mals survival problems and in our human ecosys-
from the skin of a water frog [10]. That paper could tems they grow slowly at pH 7 and their spread is
have been the first attempt to generalize, but it was inhibited by bacteria. Of course there are peptides
never followed-up and the idea was overshadowed that have evolved to be antifungal, such as those in
by the exploration of magainin derivatives as drugs. plants, but I have focused this review beginning
The first two attempts to integrate results from with insects followed by my interest in human
different animals and include antibacterial peptides peptides.
in immunology were a mini-review [11] and a Ciba There are membrane-enveloped viruses that are
Symposium with recorded discussions [12]. In susceptible to antibacterial peptides, but I believe
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
200 H. G. BOMAN

Fig. 2 Sequences for antibacterial

peptides belonging to the three
main groups. The first and the third
groups does not contain any cys-
teine residues, whilst the defensins
always have three intralinked cys-
teine residues. Note that the bond-
ing pattern differs between a- and
b-defensins. C-terminal residues
marked with an asterisk (*) are
carboxy-amidated.

this to be a minor side effect of the ability of a idin derived LL-37 (Table 1). The two first groups
particular peptide to destroy certain membranes. A contain three intramolecular disulphide bridges that
virus that cannot replicate inside a cell will disap- are typical of defensins whilst LL-37 is a linear
pear, because replication requires host ribosomes peptide without cysteine residues. It belongs to the
and energy supply. Thus, all biologically relevant cathelicidine family and is present in leucocytes,
antiviral mechanisms must provide a compromise different epithelia and testis.
between the interest of the virus and its host. In fact,
during evolution viruses may have been beneficial
Chemical synthesis and mechanisms
for species (e.g. by gene transfer), but not always for
of action
the afflicted individual.
In the Trieste database 33 entries are labelled as Before cDNA cloning became possible, the only way
‘human’. The number may seem high but both to confirm the sequence of an antibacterial peptide
cathepsin G, lactoferrin, and other precursors are was by chemical synthesis, followed by parallel
given as well as several different fragments finger printings and activity studies on different
generated by proteolysis. Some of these peptides bacteria. This was first done for cecropin A [26]. As
are just described, but not really studied in human. nature seldom is providing enough material, chem-
So far, well known human factors are the enzymes ically synthesized peptides were used to characterize
lysozyme and phospholipase. Then there are pep- the antibacterial spectra of new peptides. It still
tides: five a-defensins in neutrophils and Paneth holds that new peptides often are isolated in very
cells, four b-defensins in epithelia and the cathelic- small quantities and that synthesis is needed for the
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
 REVIEW: ANTIBACTERIAL PEPTIDES 201

Table 1 Antimicrobial peptides in

man

study of their properties. Both cecropins and the covered by peptide molecules. This saturation was
a-defensins were tested against a variety of bacteria. estimated to occur at around 1–10 billion mole-
All behaved more or less like broad-spectrum cules [30]. It is also clear that in vitro this reaction
antibiotics (tetracycline was often used as a refer- is over in a few minutes. Model membranes with
ence [21]), but as in vivo data accumulate such and without cholesterol show that cholesterol
in vitro data have become less interesting. prevents membrane damage. As this lipid is an
A few bacteria known to produce proteolytic essential component in eukaryotic membranes
enzymes, like strains of Pseudomonas, Serratia and there is a consensus that cholesterol in most cases
Staphylococcus, were generally found to be less explains why natural concentrations of antibacte-
susceptible or even resistant. Resistance to proteo- rial peptides do not cause any self-damage.
lytic enzymes could be overcome by making peptides The mechanisms of action of antimicrobial pep-
with only d-amino acids (d-enantiomers [27]). tides have recently been reviewed [31, 32]. Several
However, the method was too expensive to be useful models for explaining the collapse have been
practically. A pair of enantiomers gave CD spectra proposed, but it is unlikely that the final answer
that were mirror images of each other [27]. The is at hand. One difficulty is that the lytic action
interpretation is that native peptides form right- may not be the primary cause of death for bacteria:
handed helices, whilst those with d-amino acids the point of no return could be passed before
formed left-handed helices. Another approach to disruption of the membrane. One should remember
overcome proteolytic resistance would be to syn- that 60 years were required to understand that
thesize peptides with b-amino acids [28]. Whether apoptosis was the ultimate mechanism of action
this offers economic or biological advantages behind the killing of bacteria by penicillin and
remains to be seen. chloramphenicol.
For all linear peptides investigated it is clear that A novel approach to study the mechanism of
bacterial membranes are the main target and lysis action was recently published [33]. The core idea is
is clearly the end of the reaction. The only to study the short-term alterations in transcription
documented exception is PR-39, which stops bac- induced by sublethal concentrations of an antibac-
terial DNA synthesis [29]. There seems to be a terial peptide. Unfortunately, 11 of 26 of the
consensus: if a bacterium is exposed to a peptide, Escherichia coli genes tried have hitherto unknown
more and more peptide molecules dock on to the functions. However, the technique might become a
surface of the bacterial membrane. Finally, the powerful tool for comparing different peptides and
membrane collapses when it is completely (?) synergism as, e.g. between LL-37 and HBD-2.
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
202 H. G. BOMAN

Defensins have been made synthetically but the Table 2 Index of binding potential for peptides with antibacterial
efforts and costs have so far limited their use as a activity

tool to investigate their mechanism of action. Thus,

today, it is fair to say that for linear peptides the
bottleneck in understanding is the reaction rate and
the final collapse. However, the mechanism(s) of
action of defensins is hardly known at all. In most
cases reported claims rest on the use of a special
mutant strain of E. coli and synthetic peptide
variations have not been tried.
Chemical synthesis has been used to make a
number of designed peptides, including chimerical
peptides (called hybrids) with the basic head
(N-terminus) from cecropins and the hydrophobic Index (for protein-binding potential) is the sum of the free energies
tail (C-terminus) from an other peptide. Initially the of the respective side chains for transfer from cyclohexane to water
cecropin-melittin ‘hybrids’ were made 26 residues or taken from Radzeka and Wolfenden [37] and divided by the total
number of residues. The calculated values are negative (except for
longer, but they could be shortened down to 12–15 the hybrid peptide), but the + and ) are reversed. The per cent of
residues without loss of activity [34]. Hybrid mole- positively charged residues is designated % basic. To calculate the
cules often showed better antibiotic properties than index, we have developed a simple program named ‘Molpep 3.5’
their respective parents and one of these is included which is freely available online at http://www.ki.se/jim. The
sequences for all peptides are given in Fig. 2.
in Table 1. The sequences of peptides could be
reversed without loss of activity (retropeptides with
residues in reverse order) [35, 36]. This meant that active intestinal peptide (VIP) is a neurotransmitter
neither the direction of the peptide bond, nor the that happens to be antibacterial. All the peptides
turn of the helix (the d-enantiomers) did interfere have a positive net charge, but with threefold
with the membrane-lytic mechanism of the peptides. variations. However, the per cent of positively
The goal was to understand more about mechanism charged residues gives a more coherent picture,
of action, but occasionally there was a hope for an but it is incomplete because there are no accounts of
efficient drug. Mechanisms of action and synthetic residues with carboxyl group and those with poten-
peptides have been well covered in reviews [22, 31] tial for the weak interactions. To calculate an index
and my consensus given here is close to that of these for binding potential one could use published ther-
reviewers. modynamic solubility properties for amino acid side
Is there a common denominator for antibacterial chains [37]. The sum of the solubility values for all
peptides? In the past most authors have agreed on a residues in a sequence might give an overall
positive net charge (to facilitate binding to bacterial estimate of the potential of a peptide to bind to
phospholipids) and on an element of amphipathicity other proteins such as different receptors. To nor-
with a hinge that will help the molecule to ‘flip’ into malize such a sum it would have to be divided by the
a bacterial membrane. These criteria are rather number of residues: this was carried out for the
general and they fit groups of other polypeptides like index in Table 2. End group charges normally
histones and angiotenins, which also often have cancel, but not with a C-terminal amidation. Strictly
antibacterial activity. Therefore it is reasonable to speaking, neither proline residues nor amidations
ask: is there a way to distinguish the antibacterial are side chains and they are disregarded in Table 2.
(antimembrane) activity from the potential for For PR-39 with 18 proline residues this may be
hormone action? somewhat misleading.
An attempt to do this is presented in Table 2 in There are three peptides namely LL-37, PR-39
which the last column gives calculated values for and VIP with index values 2.48 or higher. They are
potential protein interaction (index). The sequences predicted to have a high binding potential. VIP is
for the 10 peptides are included in Fig. 2. All have regarded as a neurotransmitter and the multifunc-
antibacterial activity and some of them are also tional data for the other two would justify also
multifunctional with hormone like activities. Vaso- regarding these peptides as potential hormones.
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
 REVIEW: ANTIBACTERIAL PEPTIDES 203

The cecropin-melittin hybrid CA(1–7)M(2–9) was – but perhaps the differences between men and mice
designed only to be antibacterial and it has a is greater than often assumed by scientists.
negative index ()0.54). Of the natural peptides
magainin has the lowest index (0.42). It predicts a
cDNA cloning gave precursors
low potential for interaction with receptors and the
frog may not need any signalling from its effector. It was unexpected that it should be more difficult to
Thus, a low index value may indicate an antibac- clone smaller proteins than larger ones – but in a
terial drug candidate without many side-effects. haystack it may be easier to find a screwdriver than a
Bacterial membranes do contain ‘islands’ of pro- small screw. After 2 years, a cecropin B clone was
teins, but phospholipids with or without carbohy- found in 1985 and the cDNA sequence indicated a
drate chains are the bulk material. As the binding of minor error in the C-terminus for the peptide struc-
peptides to bacterial membranes involves the lipids ture. The peptide with an altered C-terminus was
rather than proteins, the index should be unrelated synthesized and by finger printing and mass analysis
to the bactericidal effects of the peptides. shown to be the same as the native product [40].
The index values for human defensins, HNP-1, Lehrer’s group had isolated defensins from a
HBD-2 and HBD-3 are rather similar (1.07, 0.90 number of species and they selected human defen-
and 0.90, respectively). A significant part of the sins for their first cloning that was reported in 1988
binding potential for all defensins is probably tied [41]. For all antibacterial peptides the cloning
up by their internal triple-stranded b-sheet struc- work was important because this both confirmed
ture and their ability to form dimers, all in sequences and provided conceptual news. All pep-
agreement with published solution structures for tides are made from precursor genes that also are
human b-defensins [38]. coding for a signal peptide. Moreover, for some
The index values for cecropin P1 (1.80) and peptides, post-transcriptional modifications take
CRAMP (1.59) are in the middle range. We now place at both ends. An N-terminal Glu-residue might
know that cecropin P1 is produced by the nematode become cyclicized to pyroglutamate (that blocks the
Ascaris living in the small intestine of pig and Edman reaction) whereas amidation at C-terminus is
human. It may well be able to fulfil its function obtained by an amino group from a penultimate
without much signalling. A knockout mouse in the glycine residue, followed by a basic residue. With
CRAMP gene shows that the peptide protects the this information at hand, two cecropin precursors
mice skin against infections, but no other function is were synthesized and then processed in vitro [42]. It
yet described [39]. was found that the signal peptidase requires the full
The two peptides, PR-39 and LL-37, have higher length of the signal, whilst the dipeptidyl amino-
index values (3.04 and 3.00, respectively) than the peptidase is a proline-specific enzyme. The antibac-
neurotransmitter VIP (2.53) and there are many terial activities of analogues with truncated
reports that these peptides are multifunctional. N-termini were investigated with and without pro-
LL-37 and CRAMP are both derived from cathelin cessing enzymes. It was found that the propart of the
precursors and they do have certain similarities in precursors could block the antibacterial activity.
sequence (Fig. 2), but the difference in their index The ordered cleaving of precursors is a very
values is rather large (3.00 vs. 1.59). LL-37 is important activation step, and is different for differ-
unique to humans and CRAMP is unique to mice: in ent peptides or peptide families. From the procecro-
both cases these are derived from the single cathe- pins two dipeptides were removed stepwise by a
licidin gene in the genome. Cell culture experiments dipeptidyl peptidase purified from Cecropia haemo-
may give further clues about the functional differ- lymph. The cleavage sites in the proregion were
ences between these two peptides. determined with a synthetic procecropin having a
In summary, there is a relatively good correlation proregion with two labelled proline residues with an
between the index in Table 2 and the known unlabelled residue in between. The manual Edman
properties of the respective peptides. However, for sequencing of this peptide gave radioactivity in steps
HBD-3 the index may be misleadingly low. The two and four. Using high-performance liquid chro-
difference in index between LL-37 and CRAMP was matography (HPLC) analysis during the processing
unexpected as they are two rather similar molecules we could resolve the precursor, the intermediate
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
204 H. G. BOMAN

(with one dipeptide removed) and the end product,

From defensins to epithelial protection
the mature cecropin [42]. With a larger propeptide
the same mechanism is used for the activation of the Three findings around 1990 changed the perspec-
bee venom toxin melittin (see review [43]). tive on defensins: the cloning of the a-defensins, the
finding of insect defensins and the discovery of the
b-defensins. The two defensin families are very
The first few genes and their harvest
similar in 3-D structures and in functions, in fact
At that time genomic sequences were made manu- they are almost interchangeable in nature. The
ally and in Cecropia the genes were large because the neutrophils of most mammals have granules con-
introns were more than 10 times larger than the two taining a-defensins, but bovine neutrophiles have
exons. The first gene structure solved for an anti- granule packed with b-defensins [50]. The a-defen-
bacterial peptide was the one for cecropin B in sins are made constitutively and do not seem to be
Cecropia [44]. As it was the first gene of this type, induced whilst most b-defensins investigated are
there was nothing to compare it with. We could see inducible. The a-defensins have evolved to work
that there were two exons and a large intron. inside a phagosome, they have propeptides, but in
Upstream there was a Cat box, a TATA box and a Cap the granule they are processed. If they are released
site and downstream a PolyA site but that was all. their cytotoxic properties can hurt the host. The
The cloning of two more cecropin genes and the b-defensins, like TAP in bovine airways [51], have
arrangement of a 20-kB cecropin cluster gave two signal peptides but no propeptides (made for excre-
surprises. The introns in the genes for cecropin A tion) and they seem to be produced by all epithelial
and cecropin D both contained movable retro cells. They are important for the protection of
elements, with inverted repeats (Mariner) or direct mucosa both in the lungs and in the digestive
repeats [45]. Horizontal gene transfer is becoming system [52].
accepted as part of our evolution [46], but whether Structurally, defensins are flat triple-stranded
or not cecropin genes have moved horizontally is an molecules held together by three intramolecular
open question. There may never have been selection disulphide bonds (Fig. 2). The differences between
pressure for human cecropins because 100 years the two groups are the ways the cysteine residues
ago most humans harboured two cecropin pro- are linked and the fact that the b-defensins generally
ducers. are slightly larger and some have modified termini.
One was the stomach bacterium Helicobacter pylori The 3-D structure of both types of defensins were
where the N-terminal sequence of ribosomal protein determined by X-ray diffraction and NMR. From the
L1 was found to be cecropin-like. Two synthetic first X-ray work it is clear that defensins are packed
peptides with N-terminal residues were found to be together as dimers [53]. This structure almost
potent antibacterial agents that H. pylori itself was foreboded the important finding of the h-defensins
resistant to [47, 48]. Antiserum raised against the [25]; these are small circular molecules (all peptide
synthetic peptides with residues 2–20 could identify bonded amino acids) made from two separate genes,
natural peptides in HPLC fractions of H. pylori producing two separate defensins, finally linked
extracts. In vivo, such N-terminal peptides covalently. However, whilst the a- and b-defensins
could counter-balance Lactobacteria competing are triple-stranded b-sheets, the h form is only
with H. pylori. double-stranded. So far the h defensins have only
The second case was the presence of nematodes. been found in certain primates. The digestive system
Three potent antibacterial peptides were isolated of mammals is like a fermentor for ‘steady-state’
from two parasitic nematodes Ascaris lumbricoides culturing of the natural flora of the respective
and Ascaris suum which as five instar larvae (about animals. Thus, the intestinal epithelia can be
10 cm long) live in the digestive system of humans expected to be centres for antibacterial functions.
and pigs. The pig was one of the first animals Groups working with mice and pigs realized this at
domesticated by Homo sapiens [49] and it is likely about the same time. The mouse work was carried
that some 10 000 years ago our ancestors and out in small-scale and gave first mRNA and genes
their pigs shared both natural flora and antibacterial and after 3 years the first purified intestinal defensin
peptides controlling the flora. called cryptdin [54].
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
 REVIEW: ANTIBACTERIAL PEPTIDES 205

The pig work produced results the other way An elegant study shows that intestinal defensins
round. Viktor Mutt had built a ‘factory’ for pig were found in urine of patients who had their
intestinal hormones that processed 1 metric tonne bladder replaced by an artificial one made from their
per week of frozen small intestine. In Mutt’s side own ileum [67]. Another way to demonstrate the
fractions we found first cecropin P1 [55], then functional capacity of a human gene is to insert it in
PR-39 [56] and finally three more peptides derived mice. Such transgenic animals (trangenes) often
from known proteins. Cecropin P1 was recently express several copies of their new gene, but the
found to be derived from contaminating nematodes organ-specific expression is preserved. Mice nor-
[57]. PR-39 belongs to the cathelicidins discussed in mally highly sensitive to Salmonella typhimurium
the next section. Later pig b-defensin (PBD-1) was became resistant when transfected to carry and
isolated from the tongue and the surface of this express human enteric defensin HBD-5 [68].
tissue contains the peptide in antimicrobial concen- The human skin is protected by HBD-3, which was
trations [58]. The work with porcine small intestine first isolated from psoriatic scales and cloned from
also yielded a larger peptide, NK-lysin with 78 keratinocytes [69]. The peptide is highly effective
residues and three internal S–S bonds. It is the main against S. aureus but also effective against Gram-
effector molecule of cytotoxic T and NK cells [59] negative bacteria like E. coli and P. aeuruginosa. Many
and the structure can be adopted to kill either defensins are salt-inactivated but HBD-3 is not.
bacteria or mammalian cells. When the functions An interesting finding is that isoleucine could
became clear another group isolated the corres- induce NF-jB mediated synthesis of b-defensin HBD-
ponding human molecule, named ‘granulysin’ [60]. 2 in a human epithelia cell line [70]. However, care
The cytotoxic properties of granulysin may be due to is needed in interpretation of data from cell lines or
induction of apoptosis [61]. cultured cells as illustrated by NF-jB regulation of
The solution structure of NK-lysin was solved HBD-2 [71].
with NMR [62]. It turned out that the hairpin we
had imagined had a a-helix structure between each
Cathelicidins and human life
of the disulphide bonds. As NK-lysin was produced
only by certain lymphocytes, the barrier function of In 1992–93, Zanetti et al. discovered a group of
the intestine is maintained both by different epithe- cDNA clones which indicated that a number of quite
lial cells (like the Paneth cells) and by large different peptides found in cattle and pigs were
infiltrations of lymphocytes. A loop segment from attached to a strongly conserved propiece of about
the centre of the granulysin was synthesized and 120 residues [72]. As the carrier had been discov-
showed an activity against mycobacteria, which ered before and given the name cathelin, Zanetti
was encouraging for further work [63]. called this family of molecules cathelicidins. There
The genes for human defensins are located on are two recent reviews on cathelicidins [73, 74].
chromosome 8 and the a and b-groups are adjacent The gene structures for PR-39 from the pig and
and arranged suggesting a duplication. As in vivo the human cathelin-LL-37 [75, 76] show that the
data appear on the function of different antibacterial signal peptide and the entire cathelin part are
peptides, it is becoming evident that defensins and encoded by the first three exons whilst the process-
linear peptides work in synergy [64, 65]. The ing sequences and the mature effector peptide are
importance of synergism was stressed early on by encoded by the fourth exon (Fig. 3). In pigs the
several investigators (e.g. Elsbach, Lehrer and cathelin-containing genes for protegrins and proph-
Zasloff). enins are closely linked to the gene for PR-39 [77].
Human enteric defensins are well studied. In the All three polypeptides are mapped to homologous
intestine the synthesis already starts in the foetus [1] sites on pig chromosome no. 13. The DNA
and in adults, induction of HD-5 and HD-6 (which sequences are quite similar including the intron 3,
are a-defensins) can be obtained by lipopolysaccha- facts that are consistent with an evolution by gene
ride (LPS) in Paneth cells (review [18]). The process- shuffling and horizontal transfer [78].
ing of HD-5 into two forms (residues 56–94 and 63– Several of the cathelicidins are multifunctional
94) was recently shown to be carried out by a trypsin and of clear significance for their host. The most
originating from the Paneth cells [66]. remarkable may be PR-39 from pigs. It is a
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
206 H. G. BOMAN

Control sites for transcription

factors like NF κΒ, NF-IL6 and APRF
Hypothetical site for exon shuffling

e1 e2 e3 e4
Intron 3
DNA

Coding for

KR-FA-LLGDFFRKSKE
KIGKEFKRIVQRI
KDFLRNLVPRTES

67 residues 36 residues 23 residues 39 residues

z s s s s Preproprotein

Signal
Cathelin-derived precursor

FALL-39 or LL-37
– dipeptidyl
Primary translation product
aminopeptidase
processing

Fig. 3 A cartoon showing the arrangement of the human gene for LL-37 (proFALL-39/hCAP18). The over all structure with three
conserved exons (e1–e3) is the same in all cathelicidin genes. The intron between e3 and e4 is rater similar in different animals which
facilitates recombination and exon shuffling. The variable part is exon 4, that in pig, cow and sheep can code for different effectors,
belonging to the three main groups in Fig. 2. The N-terminal residue in cathelin is pyroglutamate (Z). Thus all precursors contain this
N-terminal.

39-residue peptide with 49% proline and 24% [83]. The peptide was synthesized, found potent, and
arginine and only 11 residues of five other amino called FALL-39 after the first four N-terminal resi-
acids. Its antibacterial activity showed that it effect- dues. Chromosome mapping was done in parallel for
ively killed bacteria by stopping their DNA and both PR-39 and LL-37 [75].
protein synthesis [29]. It was also clear that PR-39 The human cathelin precursor of LL-37 has a
could pass through membranes without any appar- molecular weight of 16 034 Da, different from
ent damage. PR-39 has the highest index in Table 2. 18 kDa implied in the name hCAP-18. The process-
The molecule can induce the synthesis of syndecans ing of the precursor is not yet finalized. Our first
involved in wound healing [79] and interfere with assumption was that the molecule was cleaved after
NADPH-dependent redox reactions [80]. The regu- KR, a typical dibasic cleavage site (not yet ruled out).
lation of PR-39 has been carefully worked out [81] However, removing two N-terminal residues with a
and shows that LPS and IL-6 induce mRNA synthesis dipeptidyl peptidase gave LL-37 [76]. Nevertheless,
in isolated granulocytes. PR-39 can also induce different views have been expressed favouring
angiogenesis [82]. Several companies are now trying granular elastase, protease 3 or bacterial proteases.
to explore PR-39 and its use as a substitute for the If the intact precursor is excreted by the granulo-
heart surgery often needed after an infarct. cytes, enzymes at the site of delivery may control the
The human precursor cathelin-LL-37 (proFALL- processing and alternative sites (and enzymes) may
39, hCAP-18) was discovered independently by four offer biological advantages. In addition it is clear
different groups, but only three reports are published that certain bacteria can inactivate LL-37 by
[83–85]. The fourth group was from industry and proteolytic degradation [86], probably a disease
did not want to publish a structure already registered promoting counter defence.
in GenBank. Our contribution was a cDNA clone that The peptide or the precursor are present in
predicted a 30-residue signal, a 100-residue precur- granulocytes, keratinocytes and in certain lympho-
sor and a C-terminal 39-residue antibacterial peptide cytes and monocytes [87] as well as in epithelia such
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
 REVIEW: ANTIBACTERIAL PEPTIDES 207

as the intestine and the skin. In the colon, the Table 3 Morbus Kostmann: unknown recessive mutation in the
expression was linked to cell development [88]. In seventeenth century

addition LL-37 and human a-defensins are present

both in Vernix Caseosa and amniotic fluid and semen
Syndrome described 1956 by Rolf Kostmann. Family tree
[2, 89]. Table 1 shows that LL-37 has a high index
Diagnosed in infancy: <0.2 x 106 neutrophils mL–1 blood
and several investigators have found LL-37 to be

A maturation arrest at the promyelocyte to myelocyte stage
multifunctional [90–92] and especially to interact

Lethal – if left untreated (life expectancy <1 year)
with a receptor [93]. Shigella infections may down-
regulate the synthesis of LL-37 [94]. Semen contains
Therapy: antibiotics and rH G-CSF (daily injections)
both LL-37 and its precursor and the molecule shows
Treatment restores neutrophil counts. Patients still suffer from
recurrent infections. Severe periodontal disease as adults
affinity for a number of proteins in the semen and
may be of importance for human reproduction [89].
Here we show that Kostmann patients are missing the
antibacterial peptide LL-37
An interesting study of the multifunction of LL-37
shows that the peptide in macrophages can reduce
levels of tumour necrosis factor (TNF) previous
upregulated by LPS [95]. Patient M
LL-37 is unique to humans; it is the single
cathelicidine present in human tissue and so far is
found only in rather low concentrations. Thus,
investigations of LL-37 have required synthetic
peptides and good antibodies for analysis. These
tools were used for studies of skin disorders like
psoriasis [90] and atopic skin infections caused by S. Patient brother H

aureus [65]. The clinical importance of LL-37 is clear

especially its synergism with the b-defensin HBD-2.
Morbus Kostmann is a severe recessive disorder
caused by an unknown mutation in the seventeenth
century in northernmost Sweden, and originally
defined as an affliction found in descendants of the
family described by Kostmann in 1956 [96]. At that Fig. 4 Upper part: Dental status of Kostmann patient M in the
year 2000. X-ray photo (left) and gum status (right) of her front
time all afflicted children died from bacterial infec- teeth. The pictures were taken at 25 years of age before the
tions during their first year of life (Table 3). Homo- patient received supporting bridges. Lower part: The correspond-
zygotic parents and siblings remained healthy. ing pictures for her brother.
Kostmann patients are born with only very few
granulocytes and therapy using recombinant G-CSF and saliva of three true Kostmann patients (Fig. 5).
became possible only after 1990. Thanks to daily One Kostmann patient has been almost cured by a
injections with this cytokine a few patients with mostly rejected bone-marrow transplantation that
restored granulocyte levels are alive today [97]. restored LL-37 and its precursor. The result correla-
However, they are still prone to infections and have ted with the oral health of the patient. So far this
to live with antibiotics at hand. In addition, the might be the most striking example of the functional
morphology of granulocytes from treated Kostmann need for human antibacterial peptides.
patients are judged as ‘activated’. The G-CSF treat- The processing of LL-37 may vary with time and
ment is life saving, but severe oral problems occur as place. However, it should be noted that often there is
the patients reach adulthood (Fig. 4). much more precusor than mature effector peptide
Granulocytes and saliva from four Kostmann (as in Fig. 6). This regulated proteolysis could work
patients, one patient with a sporadic neutropenia like an antibiotic buffer which protects against an
and one with a Kostmann-like neutropenia were overdose and thus has a physiological function. As
recently investigated [3]. Levels of oxidative burst and LL-37 is consumed, the law of mass action predicts
of lactoferrin were normal, but Western blot analysis that losses are compensated by new hydrolysis of the
revealed that LL-37 was missing both in granulocytes precursor.
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
208 H. G. BOMAN

4.5 4.5 16 4.5 16 In addition, mice have a single cathelicidin with

kDa kDa kDa kDa kDa an effector peptide called cathelin-related antimicro-
Synthetic LL37 bial peptide (CRAMP). The sequences of CRAMP and
Unrelated control (G) LL-37 are rather similar (see Fig. 2). Mice with a
Related control (H)
knockout mutation in the CRAMP gene were found
to be abnormally sensitive to Streptococcal A
Related control (F)
infections [39]. It was also shown that CRAMP is
Cyclic neutropenia (W)
expressed in the salivary gland of mice and that
Kostmann BMT (N)
human saliva contains LL-37 and its precursor [98].
Sporadic Kostmann-like (J)
Kostmann (T)
Upregulation or storing of pre-made
Kostmann (M)
effectors?
Kostmann (A)
Studies of the synthesis of immunoglobulin in
Neutrophils Plasma human B-cell cultures have revealed a transcription
factor (NF-jB) regulating the gene expression of Ig
Fig. 5 Western blot analysis of the 4.5 kDa LL-37 in granulocytes
a, and the 16 kDa cathelin-LL-37 in granulocytes b, and plasma kappa light chain through a jB motif in one of the
c, from healthy individuals and Kostmann patients. To estimate introns. NF-jB is a heterodimer consisting of the
the presence of LL-37, the left part of the gel with neutrophil p65 kDa and p50 kDa subunits. The latter is
extracts was exposed 100 times longer. Granulocyte extracts
processed from a large precursor of p105, contain-
(5000 cells per well) were separated on 10–20% SDS-PAGE,
blotted onto PVDF membranes, and detected with anti-LL-37 ing ankyrin repeats. The dimer is normally present
antibody. Plasma samples (2 lL) were analysed similarly. Figures in the cytoplasm inhibited by IjB, also a molecule
4 and 5 are taken from reference 3. with ankyrin repeats. The specificity is derived from

Fig. 6 Cartoon illustrating the

similarities between different tran-
scription factors in the NF-jB fam-
ily and their inhibitors IjB
(recoloured and truncated slide of
Dan Hultmark). The Rel domain is
brownish and followed by greenish
domain mediating the nuclear
entrance after removing of the
inhibitor. In the IjB family the
ankyrin repeats are reddish indica-
ting their inhibitory role.

Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215

 REVIEW: ANTIBACTERIAL PEPTIDES 209

a lock-key mechanism where both the DNA motif receptors (TLR) (TLR-4 needs at least CD-14 as a
and the Rel domains vary. If an inhibitor IjB binds coreceptor). However, an intriguing recent finding is
to NF-jB the complex cannot enter the cell nucleus. that in mouse intestinal epithelial cells, TLR-4
Activation occurs by phosphorylation of the ankyrin recides in the Golgi apparatus rather than on the
repeats that are then digested by the proteosome. cell surface [105]. An additional complication is the
Schematic pictures of Rel molecules and their finding of several peptidoglycan recognition proteins
inhibitors are shown in Fig. 6. At the time of its (PGRPs) [106]. The genome of Drosophila encodes
discovery, the transcription factor may have been 13 PGRPs and one of them, PGRP-LC, mediates
regarded as specific. Now NF-jB-like factors are signals via the Imd pathway. Both peptidoglycan
known to be just one part of the ‘transcriptosome’, a and LPS induce in this way the synthesis of
word beginning to be used for the multi-component antibacterial peptides [14]. Four PGRPs are encoded
transcription machinery needed around the RNA in the human genome, but their role in signalling
polymerase. has not yet been analysed.
Insects have very similar ways of regulating Scientists often treat bacterial and viral infections
transcription of antibacterial peptides. In Drosophila, as one and the same phenomenon, which can hardly
at least two signalling pathways are found; these are be biologically correct. Bacteria are recognized and
genetically defined by mutants in genes required, the killed by antibacterial peptides without any need for
Toll and Imd genes, respectively [99]. Most investi- antigen presenting cells. However, there is numerous
gators still consider Toll and Imd as equally import- evidence for links from innate immunity to T cells
ant. However, the transcription factor Relish and the and to denditric cells (review in Ref. [102]).
Imd pathway are clearly needed for the expression of It must be stressed that functional peptide con-
all tested antimicrobial peptides [100, 101]. Without centrations in different mammalian tissues in most
a functional Relish gene a fly is killed by as little as a cases are too low to be analysed directly, and can
single cell of the otherwise harmless bacterium only be followed indirectly by PCR or immunological
Enterobacter cloacae. methods. Thus, you cannot know for sure if a given
The Relish activation shows clear similarities to gene is functional, because the level of expression
the human system. Rel factors, p105 and p100, may be below what can be detected. This is relevant
have compound structures similar to Relish with to the recent finding of 28 new genes for human
C-terminal IjB-like domains that are removed by b-defensins [107, 108].
processing. In mammals, the IjB-like halves of the Mature granulocytes can be induced to make more
molecules are slowly degraded by the proteosome in of a given antibacterial peptide when stores become
a constitutive manner. There may also be links depleted [81, 109]. Nevertheless, it appears that the
between innate immunity and apoptosis [14]. beauty of antibacterial peptides in mammals is their
Vertebrates have closed circulatory systems and a prevention of most bacterial infections, which can
bone marrow for the synthesis of immune cells with occur without any special signalling, as peptides in
their specific effectors. This always involves a trans- store are sufficient. Regulation could be built-in to
port problem which has to be solved. Granulocytes the normal turnover of blood cells and the tissue in
are loaded with presynthesized peptides, especially a question, to homeostasis in general and kept at a
defensins (processed) and LL-37 (as precursor). These steady level sufficient for the control of the normal
cells rapidly move to sites of infection and they have flora as well as bacterial invaders. This is possible
a short half-life. Other blood cells also contain because the size of the normal flora is so much larger
antibacterial peptides, but in lower concentrations than most infections by unwanted bacteria. If
[87]. Epithelial cells synthesize their own peptides, newcomers are eliminated within minutes, there is
LL-37 and the tissue specific b-defensins [102]. really no need for a specific signalling. Time is saved
What about the receptor concept? The idea is that and the ‘defence budget’ can be trimmed [110, 111].
binding bacterial components to a receptor on a cell But, of course, the signalling has to be there as a
surface (cultured cells) should mimic the onset of an safety device. Upregulation may be needed under
infection. The first receptor candidate in flies was special normal conditions like conception and deliv-
Toll [103, 104], previously known from Drosophila ery or when the load of invading bacteria exceeds the
development. In vertebrates there are Toll-like level balanced by normal homeostasis.
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
210 H. G. BOMAN

evolved primarily to be a defence against pathogens.

Insects, frogs, mice and men
Instead it is selected to control the natural flora that
The picture emerging of Drosophila immunity is a otherwise would constitute a threat [11, 111]. In
system with a surprising rigidity [14]. A very wide other words it is the peptide antibiotics that make
range of microorganisms triggers the same major you stay healthy. An ecological balance will always
signalling chain, the Relish/Imd-pathway. The dis- allow a few specialized bacteria (which may have no
criminatory capacity of the system seems rather other niche) to be resistant to the antimicrobial
limited. The system is also highly conserved. The peptides of a host.
Toll/Dif-pathway used during differentiation is very The evolution has selected mechanisms of survi-
similar to the signalling from the human IL-1 and val for the species, but not for the individual. For
TLR, and there are also parallels between the Imd/ some hosts infections may in the end be fatal, but
Relish-pathways and TNF-a-triggered activation of the rate of progression may be slow enough to allow
NF-jB. The PGRPs are important recognition mol- reproduction (such as in TB). Personally, I am not
ecules in Drosophila and human homologues have sure whether there is a need for ‘self and nonself-
been found that could play a similar role. recognition’ mechanisms involving bacteria. It is
Amongst lower vertebrates, different frogs are well possible to avoid self-destruction by selecting effec-
investigated [15]. Each species seems to have many tors for the most universal targets (the bacterial cell
(30 or more) antibacterial peptides, especially in the wall) that are absent in the host. To accomplish this
skin and in the intestine. Gene activation is effected goal under normal life circumstances we seem to
by NF-jB-like molecules conserved to the extent that need at least five a-defensins, three b-defensins,
commercial antibodies against human factors could LL-37 as well as lysozyme and phospholipase (the
be used. In addition, the inhibitory effects of gluco- latter two perhaps act mostly as cleaning agents).
corticoids on peptide synthesis were first found in Simplicity is always an evolutionary advantage and
frogs [112]. In addition, frogs have rather complete a peptide-enzyme based defence is compatible with
classical immune systems, but there is no clear the concept of a limited defence budget: saving of
evidence for its function against bacteria [4]. DNA as well as energy [111].
Interestingly, frogs seems to lack defensins. Of All human antibacterial peptides are hardly
course, negative results are soft facts. However, if known by now. However, since the human genome
correct, during evolution of frogs there must have is known, it would be possible to design computer
been a selection against the defensin-type of mole- searching for new basic peptides with less than 40
cules. residues and potentially forming amphipathic
Domesticated animals like cattle, sheep, pigs and a-helices. However, the search motors available today
goats have been widely used because of easy excess to are not capable of handling short polypeptides.
material. These animals all have multiple cathelici- All animals areas exposed to bacteria are protec-
dins (pigs and cattle, 10 each), but it is difficult to ted not only by endogenous antibacterial peptides,
investigate their respective functions. However, as the but also by the microbes that already live there, i.e.
need for in vivo experiments grew, the mouse came the normal microflora that is partly self-regulating.
into focus, although mouse granulocytes turned out The flora has a certain buffer capacity, but if this is
to lack a-defensins, which hampers the use of mice as exceeded the results may be very serious. Clostridium
a model for humans. Mice have many enteric defen- difficile is part of the human intestinal flora,
sins (cryptdins) which are well studied [16]. The normally only present as a limited number of spores.
peptide pattern in the small intestine of germ-free mice However, if the flora is damaged by human treat-
is very similar to animals in which the normal flora ment (like a prolonged antibiotic cure), Clostridia
had been reintroduced [113]. However, similar to spores mature and the growing bacterium produces
humans, mice have only a single cathelicidin and this a toxin that destroys the mucosal membranes giving
has been useful for in vivo experiments [39]. severe diarrhoea. So far the best treatment has been
The conservation of signalling pathways and even to remove what is left of the old flora and replace it
recognition molecules like PGRP is understandable: by the flora of a healthy nearby donor.
we meet similar bacteria and they all have pepti- Malaria infections are one of the World’s most
doglycan cell walls. The system may not have serious health problems with millions of deaths each
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
 REVIEW: ANTIBACTERIAL PEPTIDES 211

year, mainly amongst children. This was the main exposed to the same microbial environment. When it
reason for sequencing the full genome of the malaria comes to infections, the classical approach has been
mosquito Anopheles gambiae [114]. As the Dro- to cultivate the suspected pathogen. That approach
sophila genome was available and as innate immu- does not give the correct information: what we need
nity is better understood in the fly than in any other are assays for fast acting immune effectors. Immu-
animal, it was interesting to see how much of the fly nological assays with increasing sensitivity and
data would help in the understanding of the improved accuracy will probably come, but they will
mosquito [115]. only quantify what we already know to exist. Our
The life stages of the malaria parasite have long finding of LL-37 in saliva and its absence in patients
been known and many attempts have been made to with morbus Kostmann [3] indicates that saliva
make vaccines that block the liver or the red blood could be a convenient mirror of granulocyte compo-
cell forms. However, much fewer details are known nents.
about the mosquito parts of the life cycle. In the Of the antibacterial peptides already known, will
Anophiles genome there are four defensin and four some of them become new drugs? Yes, that seems at
cecropin genes that are induced by a malaria present rather likely. For known structures, modifi-
infection [116]. This finding of cecropin genes is cations may be needed in order to obtain adequate
satisfactory from a personal point of view, because patent protection; if not there is hardly any com-
one reason to start with Drosophila in 1970 was the mercial interest. There are several scientists involved
long term goal of understanding insect borne in drug development [73, 74, 119] and several
diseases. Recently a single-chain immunotoxin with peptides are in stage III of clinical trials. In a recent
a cecropin-like effector part was shown to kill review there is a table listing seven companies
ookinetes of the malaria parasite Plasmodium berghei involved in the development of antibacterial peptides
in vitro [117]. Moreover, mosquitoes with recom- as drugs [120].
binant E. coli expressing this immunotoxin in their If antibacterial peptides become available on the
gut were able to block 95.6% of the malaria market – will resistant mutants automatically
transmission in the next blood meal. become a problem? As a forecast, I would say
When an uninfected mosquito takes a blood meal ‘probably not’ and refer to the following arguments:
from a malaria-infected human, some of the para- 1 If bacterial membranes are the targets, experi-
sites enter as gametocytes. In the gut of the ments carried out indicate that it is quite difficult to
mosquito, they begin to mate and form zygotes. isolate mutants with an altered membrane composi-
The blood meal involves a very complex physiology, tion. It has been done, but the changes introduced
needed both for the mating of the parasites and for (altered charge) decreased the viability of the mutants
the egg laying of the host. It also affects the gut to such an extent that they hardly would survive in
bacterial population, which increases and peaks nature.
2 days after a blood meal [118]. A few mosquitoes 2 If the peptide structure (without any modifica-
kill all parasites in a blood meal. They cannot tions) comes from an animal, it does not involve a
transfer the parasite and are termed ‘refractory’. new selection pressure. This means that the resist-
One way to break the infectious cycle would be to ance mutations that could occur have already
alter the blood meal digestion in such a way that it happened. If a natural peptide structure is altered
always includes the killing of parasites. When that (for patent reasons), argument (1) is still valid,
goal is accomplished it remains to spread the trait in whilst (2) may be weaker.
the mosquito population either genetically or by 3 If one is to mimic nature one should always work
their gut flora or their symbionts. with a combination of peptides – that is how we stay
healthy. A resistant mutant would then have to be
altered in such a way that several peptides made
Antibacterial peptides in medicine – a
from different genes are simultaneously inactivated.
forecast
This is unlikely ever to happen.
It is medically important to understand why people My long-term guess would be that future drugs
stay healthy. Why does one person fall ill, whilst the would be designed to interfere with the synthesis of
one next to him stays well, despite both being host peptides in the organ or location desired. Such a
Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215
212 H. G. BOMAN

drug may already have been discovered, namely 11 Boman HG. Antibacterial peptides: key components needed
isoleucine. This amino acid can in cell cultures in immunity. Cell 1991; 65: 205–7.
12 Boman HG, March J, Goode J (eds). Antimicrobial Peptides.
induce the synthesis of HBD-2 [70]. If these results Ciba Symposium No. 186. Chichester, UK: John Wiley &
turn out also to be valid for human beings, it would Sons Ltd, 1991.
come close to a dream for the patient and the doctor. 13 Tossi, A. http://www.bbcm.univ.trieste.it/tossi/page2.html,
2002.
14 Hultmark D. Drosophila immunity: paths and patterns. Curr
Conflict of interest statement Opin Immun 2003; 15: 12–9.
15 Simmaco M, Mignogna G, Barra D. Antimicrobial peptides
No conflict of interest was declared. from amphibian skin: what do they tell us? Biopolymers
1998; 47: 435–50.
16 Ouellette AJ, Selsted ME. Paneth cell defensins: endogenous
Acknowledgements peptide components of intestinal host defense. FASEB J
1996; 10: 1280–9.
I thank H. Bennich and David Andreu, Charles Bevins, 17 Lehrer RI, Ganz T. Defensins of vertebrate animals. Curr Opin
Ingrid Faye and Dan Hultmark for their valuable help. Immunol 2002; 14: 96–102.
18 Huttner KM, Bevins CL. Antimicrobial peptides as mediators
Gunnar von Heine kindly checked my originally of epithelial host defense. Pediatr Res 1999; 45: 785–94.
derived index numbers. Then Eric Björwall added the 19 Ouellette AJ, Bevins C. Paneth cell defensins and innate
index calculation to Molpep 3.5. Finally Katrin Pütsep immunity of the small bowel. Inflam Bowel Dis 2001; 7: 43–
helped me to correct the final manuscript draft. 50.
20 Porter E, Bevins CL, Ghosh D, Ganz T. The multifaceted
Paneth cell. Cell Mol Life Sci 2002; 59: 156–70.
21 Boman HG. Peptide antibiotics and their role in innate
References immunity. Annu Rev Immun 1995; 13: 61–92.
1 Mallow EB, Harris A, Salzman N, Russell JP, Deberardinis RJ, 22 Andreu D, Rivas L. Animal antimicrobial peptides, an
Ruchelli E, et al. Human enteric defensins – gene structure overview. Biopolymers 1998; 47: 415–33.
and developmental expression. J Biol Chem 1996; 271: 23 Gallo RL, Huttner KM. Antimicrobial peptides: an emerging
4038–45. concept in cutanous biology. J Invest Dermatol 1998; 111:
2 Yoshio H, Tollin M, Gudmundsson G, Lagercrantz H, 739–43.
Jörnvall H, Marchini G, et al. Antimicrobial polypeptides of 24 Jack RW, Bierbaum G, Sahl HG. Lantibiotics and Related
human vernix caseosa and amniotic fluid: implications for Peptides. Berlin, Heidelberg: Springer-Verlag, 1998.
newborn innate defense. Pediatr Res 2003; 53: 211–6. 25 Tang Y, Yuan J, Osapay G, Osapay K, Tran D, Miller C. et al.
3 Pütsep K, Carlsson G, Boman HG, Andersson M. Deficiency A cyclic antimicrobial peptide produced in primate leuko-
of antibacterial peptides in patients with morbus Kostmann: cytes by the ligation of two truncated alpha-defensins. Sci-
an observational study. The Lancet 2002; 360: 1144–9. ence 1999; 286: 498–502.
4 Simmaco M, Mangoni ML, Boman A, Barra D, Boman HG. 26 Andreu D, Merrifield RB, Steiner H, Boman HG. Solid-phase
et al. Experimental infections in Rana esculenta with Aero- synthesis of cecropin A and related peptides. Proc Natl Acad
monas hydrophila: a molecular mechanism for the control of Sci USA 1983; 80: 6475–9.
the natural flora. Scand J Immunol 1998; 48: 357–63. 27 Wade D, Boman A, Wåhlin B, Drain CM, Andreu D, Boman
5 Boman HG, Nilsson I, Rasmuson B. Inducible antibacterial HG, et al. All D-amino acid containing channel-forming
defence system in Drosophila. Nature 1972; 237: 232–5. antibiotic peptides. Proc Natl Acad Sci USA, 1990; 87:
6 Hultmark D, Steiner H, Rasmuson T, Boman HG. Insect 4761–5.
immunity. Purification and properties of three inducible 28 Porter E, Weisblum B, Sh G. Mimicry of host-defense peptides
bactericidal proteins from hemolymph of immunized pupae by unnatural oligomers: antimicrobial beta-peptides. J Am
of Hyalophora cecropia. Eur J Biochem 1980; 106: 7–16. Chem Soc 2002; 124: 7324–30.
7 Steiner H, Hultmark D, Engström Å, Bennich H, Boman HG. 29 Boman HG, Agerberth B, Boman A. Mechanisms of action on
Sequence and specificity of two antibacterial proteins in- Escherichia coli of cecropin-P1 and PR-39, 2 antibacterial
volved in insect immunity. Nature 1981; 292: 246–8. peptides from pig intestine. Infect Immun 1993; 61: 2978–84.
8 Selsted ME, Brown DM, DeLange RJ, Hartwig S, Lehrer RI. 30 Steiner H, Andreu D, Merrifield RB. Binding and action of
Primary structures of MCP-1 and MCP-2, natural peptide cecropin and cecropin analogues: antibacterial peptides from
antibiotics of rabbit lung macrophages. J Biol Chem 1983; insects. Biochim Biophys Acta 1988; 939: 260–6.
258: 14485–9. 31 Rivas L, Andreu D. Cecropin-melittin hybrids. A versatile
9 Selsted ME, Harwig SS, Ganz T, Schilling JW, Lehrer RI. template in the development of membrane-active agents. In:
Primary structure of three human neutrophil defensins. Mestrina A, ed. Pore Forming Peptides and Protein Toxins,
J Clin Invest 1985; 76: 1436–9. 2002; 209–59.
10 Zasloff M. Magainins, a class of antimicrobial peptides from 32 Shai Y. Mode of action of membrane active antimicrobial
Xenopus skin: isolation, characterization of two active peptides. Biopolymers 2002; 66: 236–48.
forms, and partial cDNA sequence of a precursor. Proc Natl 33 Hong R, Shchepetov M, Weiser J, Axelsen P. Transcriptional
Acad Sci USA 1987; 84: 5449–53. profile of the Escherichia coli response to the antimicrobial

Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215

 REVIEW: ANTIBACTERIAL PEPTIDES 213

insect peptide cecropin A. Antimicrob Agents Chemother 51 Diamond G, Zasloff M, Eck H, Bressaur M, Maloy WL, Bevins
2003; 47: 1–6. CL. Tracheal antimicrobial peptide, a novel cysteine-rich
34 Wade D, Boman IA, Wahlin B, Wade D, Merrifield RB, peptide from mammalian tracheal mucosa: peptide isolation
Boman HG. Shortened cecropin-A-melittin hybrids – signi- and cloning of a cDNA. Proc Natl Acad Sci USA 1991; 88:
ficant size reduction retains potent antibiotic activity. FEBS 3952–6.
Lett 1992; 296: 190–4. 52 Bevins CL. Scratching the surface – inroads to a better
35 Merrifield RB, Juvvadi P, Andreu D, Ubach J, Boman A, understanding of airway host defense. Am J Respir Cell Mol
Boman HG. Retro and retroenantio analogs of cecropin Biol 1999; 20: 861–3.
melittin hybrids. Proc Natl Acad Sci USA 1995; 92: 3449– 53 Hill CP, Yee J, Selsted ME, Eisenberg D. Crystal structure of
53. defensin HNP-3, an amphiphilic dimer: mechanisms of
36 Merrifield EL, Mitchell SA, Ubach J, Boman HG, Andreu D, membrane permeabilization. Science, 1991; 251: 1481–5.
Merrifield RB. D-Enantiomers of 15 residue cecropin 54 Ouellette AJ, Miller SI, Henschen AH, Selsted ME. Purifica-
A-mellitin hybrids. Int J Pept Protein Res 1995; 46: 214– tion and primary structure of murine cryptdin-1, a Paneth
20. cell defensin. FEBS Lett 1992; 304: 146–8.
37 Radzeka A, Wolfenden R. Comparing the polarities of amino 55 Lee J.-Y, Boman A, Sun C, Andersson M, Jörnvall H, Mutt V,
acids: side-chain distribution coefficients between vapor et al. Antibacterial peptides from pig intestine: Isolation of a
phase, cyclohexane, 1-octanol and neutral aqueous solu- mammalian cecropin. Proc Natl Acad Sci USA 1989; 86:
tion. Biochemistry 1988; 27: 1664–70. 9159–62.
38 Schibli D, Hunter H, Aseyev V, Starner T, Wiencek J, McCray 56 Agerberth B, Gunne H, Odeberg J, Kogner P, Boman HG,
B, et al. The solution structure of the human beta-defensins Gudmundsson GH. PR-39, a proline-rich peptide antibiotic
lead to a better understanding of the potent bactericidal from pig, and FALL-39, a tentative human counterpart. Vet
activity of HBD3 against Spaphylococcus aureus. J Biol Chem Immunol Immunopathol 1996; 54: 127–31.
2002; 277: 8279–89. 57 Andersson M, Boman A, Boman HG. Ascaris nematodes
39 Nizet V, Olstake T, Lauth X, Trowbridge J, Rudisil J, from pig and human make three antibacterial peptides.
Dorschner R, et al. Innate antimicrobial peptides protects the Isolation of Cecropin P1 and two ASABF peptides. Cell Mol
skin from invasive bacterial infections. Nature 2001; 414: Life Sci 2003; 60: 599–606.
454–7. 58 Shi J, Zhang G, Wu H, Ross C, Blecha F, Ganz T, et al.
40 von Hofsten P, Faye I, Kockum K, Lee JY, Xanthopoulos KG, Porcine epithelial beta-defensin 1 is expressed in the dorsal
Boman IA, et al. Molecular cloning, cDNA sequencing, and tongue at antimicrobial concentrations. Infect Immun 1999;
chemical synthesis of cecropin B from Hyalophora cecropia. 67: 3121–7.
Proc Natl Acad Sci USA 1985; 82: 2240–3. 59 Andersson M, Gunne H, Agerberth B, Boman A, Bergman T,
41 Daher KA, Lehrer RI, Ganz T, Kronenberg M. Isolation and Sillard R, et al. NK-lysin, a novel effector peptide of cytotoxic
characterization of human defensin cDNA clones. Proc Natl t and NK cells, structure and cDNA cloning of the porcine
Acad Sci USA 1988; 85: 7327–31. form, induction by interleukin 2, antibacterial and antitu-
42 Boman HG, Boman IA, Andreu D, Li Z-q, Merrifield RB, mour activity. EMBO J 1995; 14: 1615–25.
Schlenstedt G, et al. Chemical synthesis and enzymic pro- 60 Stenger S, Hanson DA, Teitelbaum R, Dewan P, Niazi KR,
cessing of precursor forms of cecropin A and B. J Biol Chem Froelich CJ, et al. An antimicrobial activity of cytolytic T cells
1989; 264: 5852–60. mediated by granulysin. Science 1998; 282: 121–5.
43 Kreil G. Processing of precursors by dipeptidylaminopeptid- 61 Gamen S, Hanson DA, Kaspar A, Naval J, Krensky AM, Anel
ases: a case of molecular ticketing. TIBS 1990; 15: 23–6. A. Granulysin-induced apoptosis. I. Involvement of at least
44 Xanthopoulos KG, Lee JY, Gan R, Kockum K, Faye I, Boman two distinct pathways. J Immunol 1998; 161: 1758–64.
HG. The structure of the gene for cecropin B, an antibacte- 62 Liepinsh E, Andersson M, Ruysschaert, Otting G. Saposin
rial immune protein from Hyalophora cecropia. Eur J Biochem fold revealed by the NMR structure of NK-lysin. Nat Struct
1988; 172: 371–6. Biol 1997; 4: 793–5.
45 Gudmundsson GH, Lidholm D-A, Åsling B, Gan R, Boman 63 Andreu D, Carreno C, Linde C, Boman HG, Andersson M.
HG. The cecropin locus; cloning and expression of a gene Identification of an antimycobacterial domain in NK-lysin
cluster encoding three antibacterial peptides in Hyalophora and granulysin. Biochem J 1999; 344: 845–9.
cecropia. J Biol Chem 1991; 266: 11510–7. 64 Levy O, Ooi CE, Weiss J, Lehrer RI, Elsbach P. Individual and
46 Woese CR. Interpreting the universal phylogenetic tree. Proc synergistic effects of rabbit granulocyte proteins on Escheri-
Natl Acad Sci USA 2000; 97: 8392–6. chia coli. J Clin Invest 1994; 94: 672–82.
47 Pütsep K, Bränden K, Boman H, Normark S. Antibacterial 65 Ong PY, Ohtake T, Brandt C, Strickland I, Bugoniewicz M,
peptide from H. pylori. Nature 1999; 398: 671–2. Ganz T, et al. Endogenous antimicrobial peptides and skin
48 Pütsep K, Normark S, Boman HG. The origin of cecropins; infections in atopic dermatitis. NEJM 2002; 347: 1151–
implications from synthetic peptides derived from ribosomal 60.
protein L1. FEBS Lett 1999; 451: 249–52. 66 Ghosh D, Porter E, Shen B, Lee SK, Wilk D, Drazba J, et al.
49 Diamond J. Guns, Germs and Steel. The Fate of Human Soci- Paneth cell trypsin is the processing enzyme for human
eties. New York: Norton & Co., 1997. defensin-5. Nat Immunol 2002; 3: 583–90.
50 Yount NY, Yuan J, Tarver A, Castro T, Diamond G, Tran PA, 67 Porter EM, Poles MA, Lee JS, Naitoh J, Bevins CL, Ganz T.
et al. Cloning and expression of bovine neutrophil beta- Isolation of human intestinal defensins from ileal neobladder
defensins. Biosynthetic profile during neutrophilic maturation urine. FEBS Lett 1998; 434: 272–6.
and localization of mature peptide to novel cytoplasmic dense 68 Salzman N, Ghosh D, Huttner K, Paterson Y, Bevins CL. et al.
granules. J Biol Chem 1999; 274: 26249–58. Protection against enteric salmonellosis in transgenic mice

Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215

214 H. G. BOMAN

expressing a human intestinal defensin. Nature 2003; 422: 85 Cowland JB, Johnsen A, Borregaard N. hCAP-18, a cathelin/
522–6. pro-bactenecin-like protein of human neutrophil specific
69 Harder J, Bartels J, Christophers E, Schroder JM. Isolation granules. FEBS Lett 1995; 368: 173–6.
and characterization of human beta-defensin-3, a novel 86 Schmidtchen A, Frick I, Andersson E, Tapper H, Bjorck L.
human inducible peptide antibiotic. J Biol Chem 2001; 276: Proteinases of common pathogenic bacteria degrade and
5707–13. inactivate the antibacterial peptide LL-37. Mol Microbiol
70 Fehlbaum P, Rao M, Zasloff M, Anderson GM. An essential 2002; 46: 157–68.
amino acid induces epithelial b-defensin expression. Proc 87 Agerberth B, Charo J, Werr J, Olsson B, Idali F, Lindbom L,
Natl Acad Sci USA 2000; 97: 12723–8. et al. The human antimicrobial and chemotactic peptides LL-
71 Krisanaprakornkit S, Kimball J, Dale B. Regulation of human 37 and alpha-defensins are expressed by specific flympho-
beta-defensin-2 in gingival epithelial cells: the involvement cyte and monocyte populations. Blood 2000; 96: 3086–93.
of mitogen-activated protein kinase pathways, but not the 88 Hase K, Eckmann L, Leopard J, Varki N, Kagnoff MF. Cell
NF-kB transcription factor family. J Immunol 2002; 168: differentiation is a key determinant of cathelicidin LL-37/
316–24. human cationic antimicrobial protein 18 expression by
72 Zanetti M, Gennaro R, Romeo D. Cathelicidins: a novel human colon epithelium. Infect Immun 2002; 70: 953–63.
protein family with a common proregion and a variable 89 Andersson E, Sorensen O, Frohm B, Borregaard N, Egesten
C-terminal antimicrobial domain. FEBS Lett 1995; 374: 1–5. A, Malm J. Isolation of human cationic antimicrobial pro-
73 Lehrer R, Ganz T. Cathelicidins: a family of endogenous tein-18 from seminal plasma and its association with pro-
antimicrobial peptides. Curr Opin Hematol 2002; 9: 18–22. stasomes. Hum Reprod 2002; 17: 2529–34.
74 Zanetti M, Gennaro R, Skerlavaj B, Tomasinsig L. et al. 90 Frohm M, Agerberth B, Ahangari G, Stahle-Backdahl M,
Cathelicidin peptides as candidates for a novel class of Liden S, Wigzell H, et al. The expression of the gene coding
antimicrobials. Curr Pharm Des 2002; 8: 779–93. for the antibacterial peptide LL-37 is induced in human
75 Gudmundsson GH, Magnusson KP, Chowdhary BP, keratinocytes during inflammatory disorders. J Biol Chem
Johansson M, Andersson L, Boman HG. Structure of the 1997; 272: 15258–63.
gene for porcine peptide antibiotic PR-39, a cathelin gene 91 Turner J, Cho Y, Dinh NN, Waring AJ, Lehrer RI. Activities
family member: comparative mapping of the locus for the of LL-37, a cathelin-associated antimicrobial peptide of hu-
human peptide antibiotic FALL-39. Proc Natl Acad Sci USA man neutrophils. Antimicrob Agents Chemother 1998; 42:
1995; 92: 7085–9. 2206–14.
76 Gudmundsson GH, Agerberth B, Odeberg J, Bergman T, 92 Niyonsaba F, Iwabuchi K, Someya A, Hirata M, Matsuda H,
Olsson B, Salcedo R. The human gene FALL39 and pro- Ogawa H, et al. A cathelicidin family of human antibacterial
cessing of the cathelin precursor to the antibacterial peptide peptide LL-37 induces mast cell chemotaxis. Immunology
LL-37 in granulocytes. Eur J Biochem 1996; 238: 325–32. 2002; 106: 20–6.
77 Zhao C, Ganz T, Lehrer RI. Structures of genes for two 93 Yang D, Chen D, Schmidt A, Anderson G, Wang J, Wooters
cathelin-associated antimicrobial peptides: prophenin-2 and JI, et al. LL-37, the neutrophil granule- and epithelial cell-
PR-39. FEBS Lett 1995; 376: 130–4. derived cathelicidin, utilizes formyl peptide receptor-like 1
78 Gilbert W, De Souza SJ, Long M. The origin of genes. Proc (FPRL1) as a receptor to chemoattract human peripheral
Natl Acad Sci USA 1997; 94: 7698–703. blood neutrophils, monocytes, and T cells. J Exp Med 2000;
79 Gallo RL, Ono M, Povsic T, Page C, Eriksson E, Klags- 192: 1069–74.
brun M, et al. Syndecans, cell surface heparin sulfate 94 Islam D, Bandholtz L, Nilsson J, Wigzell H, Christensson B,
proteoglycans, are induced by a prolin-rich antimicrobial Agerberth B, et al. Downregulation of bactericidal peptides in
peptide from wounds. Proc Natl Acad Sci USA 1994; 91: enteric infections: a novel immune escape mechanism with
11035–9. bacterial DNA as a potential regulator. Nat Med 2001; 7:
80 Shi JS, Zhang G, Minton JE, Ross CR, Blecha F. PR-39, a 180–5.
proline-rich antibacterial peptide that inhibits phagocyte 95 Scott M, Davidson D, Gold M, Bowdish D, Hancock R. The
NADPH oxidase activity by binding to src homology 3 do- human antimicrobial peptide LL-37 is a multifunctional
mains of p47 (phox). Proc Natl Acad Sci USA 1996; 93: modulator of innate immune responses. J Immunol 2002;
6014–8. 169: 3883–91.
81 Wu H, Zhang G, Minton JE, Ross CR, Blecha F. Regulation of 96 Kostmann R. Infantile genetic agranulocytosis (agranulo-
cathelicidin gene expression: induction by lipopolysaccha- cytosis infantilis hereditaria): a new recessive lethal disease
ride, interleukin-6, retinoic acid, and Salmonella enterica in man. Acta Paediatr Scand 1956; 56: 1–78.
serovar typhimurium infection. Infect Immun 2000; 68: 97 Carlsson G, Fasth A. Infantile genetic agranulocytosis,
5552–8. morbus Kostmann: presentation of six cases from the ori-
82 Li P, Post M, Volk R, Gao Y, et al. PR39, a peptide reguator of ginal ‘Kostmann family’ and a review. Acta Paediatr 2001;
angiogenesis. Nat Med 2000; 6: 46–55. 90: 757–64.
83 Agerberth B, Gunne H, Odelberg J, Kogner P, Boman HG, 98 Murakami M, Ohtake T, Dorschner R, Gallo R. Cathelicidin
Gudmundsson GH. FALL-39, a putative human peptide antimicrobial peptides are expressed in salivary glands and
antibiotic, is cysteine-free and expressed in bone marrow saliva. J Dent Res 2002; 81: 845–50.
and testis. Proc Natl Acad Sci USA 1995; 92: 195–9. 99 Lemaitre B, Kromer-Metzger E, Michaut L, Nicolas E, Meister
84 Larrick, J., Hirata M, Balint R, Lee J, Zhong J, Wright S, M, Georgel P, et al. A recessive mutation, immune deficiency
Human CAP18: a novel antimicrobial lipopolysaccharide- (imd), defines two distinct control pathways in the Drosophila
binding protein. Infect Immun 1995; 63: 1291–7. host defense. Proc Natl Acad Sci USA 1995; 92: 9465–9.

Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215

 REVIEW: ANTIBACTERIAL PEPTIDES 215

100 Dushay MS, Åsling B, Hultmark D. Origins of immunity: 111 Boman HG. Innate immunity and the normal microflora.
relish, a compound Rel-like gene in the antibacterial defense Immunol Rev 2000; 173: 5–16.
of Drosophila. Proc Natl Acad Sci USA 1996; 93: 10343–7. 112 Simmaco M, Boman A, Mangoni ML, Mignogna G, Miele R,
101 Hedengren M, Åsling B, Dushay MS, Ando I, Ekengren S, Barra D, et al. Effect of glucocorticoids on the synthesis of
Wihlborg M, et al. Relish, a central factor in the control of antimicrobial peptides from amphibian skin. FEBS Lett
humoral, but not cellular immunity in Drosophila. Mol Cell 1997; 416: 273–5.
1999; 4: 827–37. 113 Pütsep K, Axelsson L, Boman A, Midtvedt T, Normark S,
102 Bevins C. Antimicrobial peptides as effector molecules of Boman H, et al. Germ-free and colonized mice generate the
mammalian host defense. In: Herwald H, ed. Host Response same products from enteric prodefensins. J Biol Chem 2000;
Mechanisms in Infectious Diseases. Karger: Basel, 2003; 275: 40478–82.
106–48. 114 Holt RA, Subramanian GM, Halpern A, Sutton GG, Charlab
103 Rosetto M, Engström Y, Baldari CT, Telford JL, Hultmark D. R, Nusskern DR, et al. The genome sequence of the malaria
et al. Signals from the IL-1 receptor homolog, Toll, can mosquito Anopheles gambiae. Science 2002; 173: 129–49.
activate an immune response in a Drosophila hemocyte cell 115 Zdobnv EM, von Mering C, Letunic I, Torrents D, Suyama M,
line. Biochem Biophys Res Commun 1995; 209: 111–6. Copley RR, et al. Comparative genome and proteome ana-
104 Lemaitre B, Nicolas E, Michaut L, Reichhart J-M, Hoffmann lysis of Anopheles gambiae and Drosophila melanogaster. Science
JA. The dorsoventral regulatory gene cassette spätzle/Toll/ 2002; 298: 149–59.
cactus controls the potent antifungal response in Drosophila 116 Christophides G, Zdobnov EM, Barillsas-Mury C, Birney E,
adults. Cell 1996; 86: 973–83. Blandin S, Blass C, et al. Immunity-related genes and
105 Hornef M, Frisan T, Vandewalle A, Normark S, Richter- gene families in Anophiles gambia. Science 2002; 298: 159–65.
Dahlfors A. Toll-like receptor 4 recides in the Golgi appar- 117 Yoshida S, Ioka D, Matsuoka H, Endo H, Ashii A. Bacteria
atus and colocalizes with internalized lipopolysaccaride in expressing single-chain immunotoxin inhibit malaria para-
intestinal epithelial cells. J Exp Med 2002; 195: 559–70. site development in mosquitoes. Mol Biochem Parasitol
106 Kang D, Liu G, Lundstrom A, Gelius E, Steiner H. et al. A 2001Mar; 113: 89–96.
peptidoglycan recognition protein in innate immunity con- 118 Pumpuni CB, Demaio J, Kent, Davis JR, Beier J. Bacterial
served from insects to humans. Proc Natl Acad Sci USA 1998; population dynamics in three anopheline species: the impact
195: 10078–82. on Plasmodium sporogonic development. Am J Trop Med Hyg
107 Jia, H., Schutte B, Schudy A, Linzmeier R, Guthmiller J, 1996; 54: 214–8.
Johnson G, et al. Discovery of new human beta-defensins 119 Hancock RE, Patrzykat A. Clinical development of cationic
using a genomics-based approach. Gene 2001; 263: 211–8. antimicrobial peptides: from natural to novel antibiotics.
108 Schutte B, Mitros J, Bartlett J, Walters J, Welsh M, Casavant Curr Drug Targets Infect Disord 2002; 2: 79–83.
T, et al. Discovery of five conserved beta-defensin gene 120 Zasloff M. Antimicrobial peptides of multicellular organisms.
clusters using a computational search strategy. Proc Natl Nature 2002; 415: 389–95.
Acad Sci USA 2002; 99: 2129–33.
109 Tomasinsig L, Scocchi M, Di Loreto C, Artico D, Zanetti M. Received 1 January 2003; revision received 22 April 2003;
Inducible expression of an antimicrobial peptide of the accepted 4 July 2003.
innate immunity in polymorphonuclear leukocytes. J Leukoc
Biol 2002; 72: 1003–10.
Correspondence: Professor Hans G. Boman, Microbiology and
110 Boman HG. Gene-encoded peptide antibiotics and the con-
Tumor Biology Center, Box 280, Karolinska Institutet, S-171 77
cept of innate immunity: an update review. Scand J Immunol
Stockholm, Sweden. (e-mail: [email protected]).
1998; 48: 15–25.

Ó 2003 Blackwell Publishing Ltd Journal of Internal Medicine 254: 197–215