Thrombocytes Are Effectors of the Innate Immune System Releasing

Injury, Int. J. Care Injured 42 (2011) 682–686

Contents lists available at ScienceDirect

Injury

jo urnal homepage: www.elsevier.com/locate/injury

Thrombocytes are effectors of the innate immune system releasing

§ human beta defensin-3

b a c b

Mersedeh Tohidnezhad , Deike Varoga , Rainer Podschun , Christoph Jan Wruck ,

a b b,1 a,1,

Andreas Seekamp , Lars-Ove Brandenburg , Thomas Pufe , Sebastian Lippross *

Department of Trauma Surgery, University Hospital of Schleswig Holstein, Campus Kiel, Germany

Department of Anatomy and Cell Biology, RWTH Aachen University, Germany

Institute for Infection Medicine, University Hospital of Schleswig Holstein, Campus Kiel, Germany

A R T I C L E I N F O A B S T R A C T

Article history: Background: Thrombocyte concentrate i.e. platelet-rich plasma (PRP) has become a popular adjunct for

Accepted 16 December 2010

many surgical procedures. It is believed to improve bone and soft tissue healing. Recently antimicrobial

effects of the autologous preparation were reported by several groups. In this study we investigated the

Keywords: antimicrobial effect of PRP against gram-negative microbes which frequently cause severe complications

Platelet-rich plasma (PRP)

in orthopaedic trauma surgery.

Pseudarthrosis

Methods: Platelet-rich plasma was produced from liquid preserved thrombocyte concentrates. ELISA,

Fracture healing

Western blot and immunohistochemistry were preformed to investigate the release and content of

Antimicrobial peptide (AMP)

platelet concentrates. A radial diffusion assay was used to detect antimicrobial effects of PRP.

Results: We detected the human beta defensin-3 in bactericidal concentrations in platelet preparations

by ELISA, Western blot and immunohistochemistry. In antimicrobial testing we demonstrated effective

inhibition of Escherichia coli (ATCC 11303), Bacterium megaterium (ATCC 14581), Klebsiella pneumoniae

(ATCC 13883), Enterococcus faecalis (ATCC 29212) and Proteus mirabilis (ATCC21100).

Conclusion: With this study we demonstrate antimicrobial action of a popular adjunct for orthopaedic

and trauma surgery against gram-positive and gram-negative bacteria. We have identiﬁed a possible

mechanism of action via the secretion of HBD-3 as a ﬁrst line defence in contaminated wounds and in

elective application of PRP. This ﬁnding supports a broader spectrum of clinical indications for an

autologous platelet preparation.

Introduction regulating the cellular events that follow tissue damage. They

adhere, aggregate, form a ﬁbrin mesh and subsequently release a

Autologous thrombocyte concentrates also termed platelet-rich large variety of growth factors and cytokines. Platelet concentrates

plasma (PRP) have gained great popularity as an adjunct to a are believed to improve tissue healing by producing higher than

variety of surgical procedures. In the beginning of this century, the physiologic concentrations of locally stimulating substances.

clinical application of PRP has been considered a breakthrough in Platelet concentrates can be prepared from whole blood within

the stimulation and acceleration of bone and soft tissue healing. a short time using relatively simple methods. Therefore the idea to

Platelets and the growth factors they release are essential for use them as an immunogenically inert additive in cases where

rapid healing and tissue regeneration is required seems self-

evident. Several companies have developed kits and devices for the

automatic preparation i.e. during surgical procedures. Up to now

This work was supported by grants of the Deutsche Forschungsgemeinschaft

PRP has been used predominantly in maxillofacial surgery as an

(SFB 627, Projekt A22, DFG Va220/21, Pu214/3-2, Pu214/4-2 and Pu214/5-2). 1

autologous additive to bone grafts and soft tissue transplants.

* Corresponding author at: University Hospital of Schleswig-Holstein, Campus

Other indications that have been proposed range from chronic

Kiel, Department of Trauma Surgery, Arnold-Heller-Strasse 3, D-24118 Kiel,

Germany. Tel.: +49 431 5974581; fax: +49 431 5973883. diabetic ulcers over the treatment of tendinitis to standard

2,4,7,9

E-mail addresses: [email protected] (M. Tohidnezhad), orthopaedic procedures. Recently a new aspect of thrombo-

[email protected] (D. Varoga), [email protected]

cyte action in clinical applications was proposed when PRP was

(R. Podschun), [email protected] (C.J. Wruck), [email protected] 6

shown to provide antimicrobial properties in vitro. In previous

(A. Seekamp), [email protected] (L.-O. Brandenburg), [email protected]

(T. Pufe), [email protected] (S. Lippross). studies we have detected antimicrobial molecules that are

These authors contributed equally to the work. produced and secreted by various tissues. The so called

M. Tohidnezhad et al. / Injury, Int. J. Care Injured 42 (2011) 682–686 683

antimicrobial peptides (AMP) are key elements of the innate Plate diffusion test

immune system providing the ﬁrst line of defence in many

13,14,28

epithelial tissues against invading microbes. Recent inves- In vitro susceptibility to PRP was investigated by a standard

tigations of these widely distributed effectors demonstrate disc diffusion test. An underlay agar (agarose: 1%, Tween 20:

expression and regulation in mesenchymal cells and tis- 0.02%, tryptic soy broth: 1%, 0.01 M phosphate buffer) and an

19,20,22–24

sues. Defensins are an important subfamily of AMP and overlay agar (agarose: 1%, caseine peptone: 3.4%, tryptic soy

kill microbes by destroying their cell membranes without use of peptone: 0.6%, glucose: 0.5%, NaCl: 1%, K2HPO4: 0.5%) were used.

the adaptive immune system. They are a complex group of 4– Agar plates were coated with Escherichia coli (ATCC 11303),

5 kDa, open-ended, cysteine-rich, cationic peptides that are Bacterium megaterium (ATCC 14581), Klebsiella pneumoniae (ATCC

divided into a- and b-defensins based on the location and 13883), Enterococcus faecalis (ATCC 29212) and Proteus mirabilis

connectivity of six conserved cysteine residues. A new antimicro- (ATCC21100). Strains were incubated overnight in tryptic soy

bial peptide, human beta-defensin-3 (HBD-3), was recently broth under shaking, inoculated into fresh medium and grown to

identiﬁed. HBD-3 was shown to have potent killing activity log-phase for 3 h at 37 8C. Bacterial suspensions (50–200 ml) were

against Staphylococcus aureus and multiple, multiresistant mixed with 10 ml underlay agar at 46 8C and poured into Petri

bacterial strains. dishes. After the agar had cooled down, 10-ml aliquots of each

The importance of antimicrobial peptides in mammalian host sample to be tested were pipetted into wells that had been

defence becomes evident in mice that overexpress the human previously punched (2.5-mm diameter) into the agar. After

HBD-5 gene and are subsequently resistant to the oral application overnight incubation at 37 8C, 10 ml of overlay agar (46 8C) was

of Staphylococcus typhimurium. The expression of antimicrobial poured onto the underlay agar. Agar plates were incubated at

peptides after bacterial contact therefore seems to be a pivotal 37 8C for 3–4 h and examined for growth inhibition zones. Each

mechanism in the prevention of infective diseases in bone. Innate test was performed in duplicate. Nine microliters of PRP was

defence mechanisms are perhaps more important than the applied and activated by adding 1 ml of thrombin (10 U/ml). Full

adaptive immunity in avascular tissue or tissue with long diffusion blood served as a control.

distances as in cartilage, bone or the superﬁcial zone of the skin.

The restricted access of adaptive immune cells into the highly HBD-3 ELISA

specialised extracellular matrix of these tissues explains the

presence of a cell-based anti-infectious host response. Following activation, the samples were centrifuged at 18,000 g

In orthopaedic trauma surgery open fractures challenge the for 2 min to pellet debris and the gel that formed after activation. The

surgeon because they are associated with a high rate of post- resulting supernatant was diluted 1:2 in PBS containing 0.1% BSA

11,15

traumatic soft tissue infections and osteomyelitis. Further- and human HBD-3 protein content was measured (HBD-3 ELISA

more open fractures and concomitant infection is the leading Development Kit Cat.Nr:900-K210, Peprotech, Hamburg, Germany).

12,17

cause for non-union of the long bones. In many of these For this examination, 96-well immunoplates (Corning Inc., New

tragic cases limb salvaging procedures fail and the injury leads York, USA) were coated at 4 8C for 24 h with 100 ml (300 mg/ml) of

to the loss of a limb after a long time of suffering. The open anti-HBD-3 antibodies diluted to 1:100 in phosphate buffer at pH

wound with vast tissue damage, heavily compromised perfu- 7.2. The wells were then washed three times with PBS + 0.1% Tween

sion, necrotic tissue and foreign material contamination 20 and blocked with 300 ml 1% bovine serum albumin in PBS for

provides ideal growing conditions for a high load of harmful 60 min at room temperature. After washing three times with 200 ml

bacteria that are naturally omnipresent. With respect to these PBS + 0.1% Tween 20, 100 ml of standard and sample per well were

unfavourable premises one can even regard the overall infection incubated for at least 1 h at room temperature. Plates were washed

rate of open fracture relatively low. The initiative salvage three times with PBS + 0.1% Tween 20 and the wells were incubated

mechanism against progressing tissue damage is mediated by for 2 h at room temperature with 100 ml of biotinylated anti-hBD-3

thrombocyte adherence with subsequent clot formation and antibodies diluted 1:100 to 0.25 mg/ml in PBS + 0.1% Tween 20.

release of locally active inﬂammatory mediators. We postulate Plates were washed again three times with PBS + 0.1% Tween 20 and

that in concert with various bioactive proteins the human beta- ﬁlled with 50 ml of streptavidin–peroxidase in each well (Roche

defensin-3 is released by thrombocytes as a local antibacterial Diagnostics, Mannheim, Germany; 1:10.000 in PBS + 0.1% Tween

agent. The recent ﬁnding of PRP possessing antimicrobial 20). The plates were then incubated for 30 min at room temperature,

activity in vitro supports this thesis. In this study we washed three times as described above, and incubated with 100 ml

0 0

investigated the secretion of antimicrobial peptides by throm- 3,3 ,5,5 -tetramethylbenzidine (TMB; Sigma, Taufkirchen,

bocytes in order to elucidate the mechanism of thrombocyte Germany), which served as the developmental agent, for a

anti-infective capabilities. maximum of 15 min in the dark at room temperature followed by

100 ml of 1 M H2SO4. Absorbance was measured at 450 nm with a

Materials and methods multichannel photometer (Sunrise; Tecan, Crailsheim, Germany).

The lower and upper detection limits were 1–4000 ng/ml. Human

Preparation of PRP recombinant HBD-3 served as the standard for the following

concentrations: 0, 0.06, 0.125, 0.25, 0.5, 1.0, 2.0 and 4.0 ng/ml.

Platelet rich plasma was produced from liquid-preserved

platelet concentrates obtained by plateletpheresis that were not Western blot

older than one day, in accordance with the current ethical laws of

Germany. The Plasma contained 1.9 10 platelets per ml and a For Western blot, samples were reduced in the presence of

maximum of 2400 leucocytes per ml. The plasma was concentrated 10 mM dithiothreitol, proteins were separated by sodium dode-

2-fold by centrifuging at 2000 g for 5 min to achieve 5- to 10-fold cylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE; 15%

blood concentration. The remaining supernatant was preserved as gels), transferred onto nitrocellulose membranes that were blocked

platelet-poor plasma (PPP). PRP was activated using bovine and incubated with antibodies according to standard techniques

thrombin (10 U/ml) (Sigma, Germany) at 37 8C for 30 min in the (HBD-3: Santa Cruz, Cat-Nr:sc-30115, Heidelberg, Germany).

presence of 2.5 mM CaCl2. Both were stored at 20 8C Signals were detected by chemoluminescence reaction (ECL-Pus;

until thawed. Amersham-Pharmacia, Uppsala, Sweden).

684 M. Tohidnezhad et al. / Injury, Int. J. Care Injured 42 (2011) 682–686

Immunohistochemistry Statistical analysis

HBD-3 expression was analysed in supernatants of PRP. 50,000 Differences between the groups were evaluated using the t-test.

thrombocytes were seeded on microscope slides for immunocy- Group differences were considered signiﬁcant if p was less than 0.05.

tochemical analyses. Slides were blocked with 3% hydrogen SPSS 13.0 software was used.

peroxide to inhibit endogenous peroxidases. They were then

incubated with normal serum (1:5 in Tris-buffered saline) from

the species in which the secondary antibody was attained. Results

Immunocytochemical staining was performed using polyclonal

primary antibodies against human b-defensin-3 (HBD-3, 1:40, Immunohistochemistry

Santa Cruz Biotechnology, Santa Cruz, USA). This was followed by

incubation with biotinylated secondary antibodies and a peroxi- HaCaT ceratinocytes served as positive control and revealed a

dase-labelled streptavidin–biotin staining technique (DAKO, strong staining for HBD-3 as expected (Fig. 1a). The negative

Glostrup, Denmark). After counterstaining with hemalum, the control was performed by omitting the primary antibody

sections were mounted with Aquatex (Boehringer, Mannheim, (Fig. 1b). Immunohistochemistry revealed strong positive

Germany). Negative controls were performed by omitting the staining for HBD-3 in thrombocytes (Fig. 1c).

primary antibody or by pre-incubation of the primary antibody

with recombinant human hBD-3 (100 mg/ml). Western blot

As positive control HaCaT ceratinocytes were used. The cells

were incubated on cover slips for 36 h, washed with PBS and Western blot analysis veriﬁed these ﬁndings (Fig. 2). The band

ﬁxed using Zamboni for 30 min. They were washed for three for HBD-3 was veriﬁed using recombinant HBD-3 at 5 kDa. T1–T3

times and stored at 4 8C with saturated humidity till represent the supernatant of thrombocytes from different donors

staining procedure. after thrombin treatment.

Fig. 1. Immunohistochemical staining of thrombocytes against HBD-3: (a) positive control by staining of HaCaT (ceratinocytes); (b) negative control was performed by

omitting the primary antibody; (c) human thrombocytes revealed a strong staining for HBD-3; (d) HBD-3 additionally detected in plasma. Bar represents 20 mm, nuclei counterstained with hemalum.

M. Tohidnezhad et al. / Injury, Int. J. Care Injured 42 (2011) 682–686 685

(ATCC21100) by use of a plate diffusion agar assay. This assay

was used to compare antimicrobial activities against gram

positive and gram negative bacteria of PRP to whole blood. As

shown in Fig. 4 PRP showed a larger diameter of the inhibition

Fig. 2. Western blot against human beta defensin-3 in thrombocytes: detection of

zone against E. coli, B. megaterium, K. pneumoniae, E. faecalis and P.

HBD-3 in supernatant of thrombin activated thrombocytes (T). T1–T3 represent

mirabilis (5.3 1.1, p < 0.015, 10.5 1.1, p < 0.036, 5.0 0.7,

probes from different donors. Recombinant human beta defensin-3 served as

positive control. Standard and supernatant revealed as estimated bands at 5 kDa. p < 0.008, 3.5 0.1, p < 0.001, and 6.0 0.7 mm, p < 0.173, respec-

tively) and therefore higher activity against the test organisms than

full blood. The results were signiﬁcantly higher for PRP than for full

blood with exception of P. mirabilis.

Discussion

Bone and soft tissue infections are feared as disastrous

complications of open fractures in elective orthopaedic surgery.

The problem is well recognised amongst practising surgeons and

therefore some speciﬁc treatment principles are well respected

today. Damage control surgery comprises early and radical

debridement of open fractures and deep wound infections,

external ﬁxation if the soft tissue is heavily compromised, and

3,18

antibiotic cover. By these means the rate of infections after

second degree open fractures today ranges from 1% to 55%

depending on the severity of the injury according to various

authors. Although this rate seems high compared to closed

fractures in a considerable number of cases severely contaminated

tissue heals without complications. A local immediately acting

Fig. 3. ELISA: HBD-3 was quantiﬁed in PPP and PRP supernatant using ELISA. HBD-3

pathomechanism could represent a ﬁrst line defence for antimi-

amounts in PRP supernatant were signiﬁcantly higher compared to PPP

supernatants (p < 0.001). crobial invasion that is effective in these cases.

With this study we provide evidence that thrombocytes could

ELISA be the effectors of such a rapidly available defence mechanism by

releasing the antimicrobial peptide HBD-3.

The levels of HBD-3 in PRP and PPP were quantiﬁed using Platelets can be considered ideal candidates for the delivery of

ELISA. HBD-3 concentration in PPP was 2845.4 1781.2 pg/ml. bioactive substances because their adherence to damaged

HBD-3 concentration in activated PRP was 6146.3 944.4 pg/ml. endothelium is the ﬁrst response to any kind of injury. Ahead of

HBD-3 level in PRP was signiﬁcantly higher than in PPP all cellular action platelets mechanically form a mesh for

(p < 0.001;n = 14) (Fig. 3). regenerative cells to adhere and release a vast amount of growth

factors and inﬂammatory mediators. By stopping bleeding and

Antimicrobial activity of PRP stimulating regenerative capacities the organism initiates salvage

of damaged tissue and inhibits further progression of the damage.

The antimicrobial activity of PRP was evaluated against E. coli To make use of the potent combination of a jellifying liquid and a

(ATCC11303), B. megaterium (ATCC14581), K. pneumoniae large amount of regenerative molecules in surgery platelet-rich

(ATCC13883), E. faecalis (ATCC29212) and P. mirabilis plasma was introduced as an autologous concentrate of thrombo-

cytes in plasma. Today PRP is mainly used in cases of non-union

and to augment fragile reconstructions in maxillofacial surgery

hence with increasing evidence of anti-infective capabilities a new

ﬁeld for the application is conquered. The molecular mechanism of

action of HBD-3 has been thoroughly investigated in the past.

Our analysis of PRP-supernatant revealed substantial HBD-3

content in the range of 5–7 mg/ml. This concentration has

previously been shown to be biologically effective for the

inhibition of bacterial growth. In due course we investigated

PRP for speciﬁc antibacterial action by plate diffusion testing.

Many authors have demonstrated that HBD-3 is a potent agent

against gram positive and gram negative bacteria. PRP as the

source of HBD-3 inhibits gram positive growth as recently reported

by Bielecki et al. We present a novel aspect by showing that PRP

can inhibit gram-negative E. coli, E. faecalis, B. megaterium and K.

pneumoniae in addition to gram positive organisms.

This feature of a platelet concentrate is of special interest with

respect to the fact that gram negative microbes are frequently

found as open fracture contaminating bacteria. In one study PRP

was used for the treatment of leg ulcers that continuously cause

problems with persisting superinfections. In such cases, too, can

Fig. 4. Plate diffusion test: platelet-rich plasma revealed effective inhibition of

gram negative microbes be the pathogen that might be targeted by

growth for E. coli, B. megaterium, K. pneumoniae, E. faecalis and P. mirabilis (5.3 1.1,

PRP-derived antimicrobial action. Additionally positive effects of a p < 0.015, 10.5 1.1, p < 0.036, 5.0 0.7, p < 0.008, 3.5 0.1, p < 0.001, and

6.0 0.7 mm, p < 0.173, respectively). platelet-leucocyte gel on healing of delayed union were described.

686 M. Tohidnezhad et al. / Injury, Int. J. Care Injured 42 (2011) 682–686

References

The aetiology of delayed union and non union is at least partly

believed to be infective.

1. Anitua E, Andia I, Ardanza B, et al. Autologous platelets as a source of proteins for

As such we demonstrate two independent novel ﬁndings—

healing and tissue regeneration. Thromb Haemost 2004;91:4–15.

extended bacterial growth inhibition by PRP and considerable 2. Aspenberg P, Virchenko O. Platelet concentrate injection improves Achilles

tendon repair in rats. Acta Orthop Scand 2004;75:93–9.

content of the HDP HBD-3 in PRP. However with this study we

3. Bhandari M, Guyatt GH, Tornetta III P et al. Current practice in the intrame-

do not provide data clearly showing a mechanistic association of

dullary nailing of tibial shaft fractures: an international survey. J Trauma

HBD-3 content and antimicrobial effects of PRP. Still, as we have 2002;53:725–32.

4. Bhanot S, Alex JC. Current applications of platelet gels in facial plastic surgery.

investigated the action of HDP in various circumstances before

Facial Plast Surg 2002;18:27–33.

we want to point out some supporting ﬁndings. Coming from

5. Bielecki T, Gazdzik TS, Szczepanski T. Beneﬁt of percutaneous injection of

the investigation of antibiotic-resistant bacteria Yeaman et al. autologous platelet-leukocyte-rich gel in patients with delayed union and

have shown a similar mechanism for the rat. They demon- nonunion. Eur Surg Res 2008;40:289–96.

6. Bielecki TM, Gazdzik TS, Arendt J, et al. Antibacterial effect of autologous platelet

strated that platelet microbicidal protein-1, a peptide that is

gel enriched with growth factors and other active substances: an in vitro study. J

released by thrombocytes, showed inhibition of bacterial growth

Bone Joint Surg Br 2007;89:417–20.

upon activation by thrombin. Meanwhile several cationic 7. Bose B, Balzarini MA. Bone graft gel: autologous growth factors used with

25 autograft bone for lumbar spine fusions. Adv Ther 2002;19:170–5.

peptides have been detected in the rat and the antimicrobial

26 8. Carsenti-Etesse H, Doyon F, Desplaces N, et al. Epidemiology of bacterial infec-

spectrum was extensively illustrated. Further support for our

tion during management of open leg fractures. Eur J Clin Microbiol Infect Dis

hypothesis comes from our own studies on serum of multiply 1999;18:315–23.

9. Carter CA, Jolly DG, Worden Sr CE et al. Platelet-rich plasma gel promotes

injured patients (data submitted for publication). Serum of

differentiation and regeneration during equine wound healing. Exp Mol Pathol

multiply injured patients contains elevated levels of antimicro- 2003;74:244–55.

bial peptides human beta-defensin-2, -3 and the cathelicidin LL- 10. Dellinger EP. Antibiotic prophylaxis in trauma: penetrating abdominal injuries

and open fractures. Rev Infect Dis 1991;13(Suppl. 10):S847–57.

37 and likewise exhibits an increased bacteriostatic effect

11. Gustilo RB, Anderson JT. JSBS classics. Prevention of infection in the treatment

compared to control individuals. Finally these ﬁndings provide a

of one thousand and twenty-ﬁve open fractures of long bones. Retrospective

possible explanation for antimicrobial features of full blood. In and prospective analyses. J Bone Joint Surg Am 2002;84-A:682.

12. Gustilo RB, Mendoza RM, Williams DN. Problems in the management of type III

our study we used full blood as a control—in antimicrobial

(severe) open fractures: a new classiﬁcation of type III open fractures. J Trauma

testing the effect was visible but considerably weaker than that

1984;24:742–6.

of PRP. Still, a lower concentration of platelets could release 13. Harder J, Bartels J, Christophers E, Schroder JM. A peptide antibiotic from human

HBD-3 in antibacterial doses. skin. Nature 1997;387:861.

14. Harder J, Bartels J, Christophers E, Schroder JM. Isolation and characterization of

With this study we provide evidence of gram positive and

human beta-defensin-3, a novel human inducible peptide antibiotic. J Biol Chem

gram-negative antimicrobial action of a popular adjunct for 2001;276:5707–13.

orthopaedic and trauma surgery. We have identiﬁed hints for a 15. Kelly PJ, Fitzgerald Jr RH, Cabanela ME, et al. Results of treatment of tibial and

femoral osteomyelitis in adults. Clin Orthop Relat Res 1990:295–303.

possible mechanism of action via the secretion of HBD-3 as a

16. Maisetta G, Batoni G, Esin S, et al. In vitro bactericidal activity of human beta-

ﬁrst line defence in contaminated wounds and in elective

defensin 3 against multidrug-resistant nosocomial strains. Antimicrob Agents

application of PRP. We and others have started using PRP as an Chemother 2006;50:806–9.

17. Pan KL, Shukur MH, Ghani MA. Locked intramedullary nailing for complex non-

additive for the revision of non-union because it is safe and

union of the tibia. Injury 1994;25:655–7.

affordable. With our in vitro data we add information towards a

18. Pape HC, Hildebrand F, Pertschy S, et al. Changes in the management of femoral

scientiﬁc rationale for a clinical application. With respect to a shaft fractures in polytrauma patients: from early total care to damage control

orthopedic surgery. J Trauma 2002;53:452–61.

raised interest and attention to cost effectiveness of surgical

19. Paulsen F, Pufe T, Conradi L, et al. Antimicrobial peptides are expressed and

treatment this aspect gains importance. Further investigations

produced in healthy and inﬂamed human synovial membranes. J Pathol

are necessary to ultimatively prove the proposed mechanism. At 2002;198:369–77.

20. Paulsen F, Pufe T, Petersen W, Tillmann B. Expression of natural peptide

least of equal importance seem large scale trials to obtain

antibiotics in human articular cartilage and synovial membrane. Clin Diagn

profound clinical data concerning the range of indications for

Lab Immunol 2001;8:1021–3.

PRP. Along with such clinical investigations can insights be 21. Salzman NH, Ghosh D, Huttner KM, et al. Protection against enteric salmonel-

obtained as if an antimicrobial mechanism can give rise to a losis in transgenic mice expressing a human intestinal defensin. Nature

2003;422:522–6.

frequent clinical application i.e. for the prevention of infectious

22. Varoga D, Paulsen F, Kohrs S, et al. Expression and regulation of human beta-

pseudarthrosis and in revision surgery.

defensin-2 in osteoarthritic cartilage. J Pathol 2006;209:166–73.

23. Varoga D, Pufe T, Mentlein R, et al. Expression and regulation of antimicrobial

peptides in articular joints. Ann Anat 2005;187:499–508.

Conﬂict of interest

24. Varoga D, Wruck CJ, Tohidnezhad M, et al. Osteoblasts participate in the innate

immunity of the bone by producing human beta defensin-3. Histochem Cell Biol

The experiments comply with the current laws of Germany. The 2009;131:207–18.

25. Weksler BB, Nachman RL. Rabbit platelet bactericidal protein. J Exp Med

authors conﬁrm that no conﬂict of interest exists.

1971;134:1114–30.

26. Yeaman MR. The role of platelets in antimicrobial host defense. Clin Infect Dis

Acknowledgments 1997;25:951–68.

27. Yeaman MR, Puentes SM, Norman DC, Bayer AS. Partial characterization and

staphylocidal activity of thrombin-induced platelet microbicidal protein. Infect

We thank S. Echterhagen, C. Jaeschke, M. Nicolau, M. Pradella, A.

Immun 1992;60:1202–9.

Ru¨ ben, L. Shen for their excellent technical assistance and A. 28. Zasloff M. Antimicrobial peptides of multicellular organisms. Nature

2002;415:389–95.

Drapkin for the proofreading.