Downloaded from on January 9, 2020 by ... · Establishment of mouse osteomyelitis model 4 63 INTRODUCTION 64 One of the most serious problems in the orthopaedic field is infectious

Establishment of mouse osteomyelitis model

1

1

Title: 2

Establishment of a Real-time, Quantitative and Reproducible Mouse Model of 3 Staphylococcal Osteomyelitis using Bioluminescence Imaging 4

5 Authors: 6 Haruki Funao1, Ken Ishii1, Shigenori Nagai2,3, Aya Sasaki4, Tomoyuki Hoshikawa5, 7 Mamoru Aizawa5, Yasunori Okada4, Kazuhiro Chiba1, Shigeo Koyasu2, Yoshiaki 8 Toyama1, Morio Matsumoto1 9 10 Affiliations: 11 1 Department of Orthopaedic Surgery, School of Medicine, Keio University, Shinjuku, 12 Tokyo, JAPAN 13 2 Department of Microbiology and Immunology, School of Medicine, Keio University, 14 Shinjuku, Tokyo, JAPAN 15 3 Core Research for Evolutional Science and Technology (CREST), Japan Science and 16 Technology Agency (JST), Tokyo, Japan 17 4 Department of Pathology, School of Medicine, Keio University, Shinjuku, Tokyo, 18 JAPAN 19 5 Department of Applied Chemistry, School of Science and Technology, Meiji 20 University, Ikuta, Kanagawa, JAPAN 21

22 Corresponding Author’s address: 23 Ken Ishii, MD, PhD 24 Department of Orthopaedic Surgery, School of Medicine, Keio University, 25 35 Shinjuku, Tokyo, 160-8582, JAPAN 26 Phone: +81-3-5363-3812 27 Fax: +81-3-3353-6597 28 Email: [email protected] 29 30 Key words: mouse, osteomyelitis, bioluminescence imaging, in vivo imaging, 31 staphylococcus aureus 32 33 Running title: Establishment of mouse osteomyelitis model 34 35 Conflict of interest statement: S.K. is a consultant for Medical and Biological 36 Laboratories, Co. Ltd. The authors otherwise have no financial conflicts of interest. 37

Copyright © 2011, American Society for Microbiology and/or the Listed Authors/Institutions. All Rights Reserved.Infect. Immun. doi:10.1128/IAI.06166-11 IAI Accepts, published online ahead of print on 21 November 2011

on February 17, 2020 by guest

http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


2

38 39 Sources of Support: This work was supported by a Grant from the Japan Orthopaedics 40 and Traumatology Foundation, Inc. No.192, Keio Gijuku Academic Development 41 Funds, The General Insurance Association of Japan, Health Labour Sciences Research 42 Grant and Research for Promoting Technological Seeds. 43 44 Certification: I certify that this work has not been previously published. 45


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


3

ABSTRACT 46

Osteomyelitis remains a serious problem in the orthopedic field. There are only 47

a few animal models in which the quantity and distribution of bacteria can be 48

reproducibly traced. Here, we established a real-time quantitative mouse model of 49

osteomyelitis using bioluminescence imaging (BLI) without sacrificing the animals. A 50

bioluminescent strain of Staphylococcus aureus was inoculated into the femur of mice. 51

The bacterial photon intensity (PI) was then sequentially measured by BLI. Serological 52

and histological analyses of the mice were performed. The mean PI peaked at 3 days 53

and stable signals were maintained for over 3 months after inoculation. The serum 54

levels of interleukin-6, interleukin-1β, and C-reactive protein were significantly higher 55

in the infected mice than in the control mice on day 7. The serum monocyte chemotactic 56

protein-1 level was also significantly higher in the infected group at 12 hours compared 57

to the control group. A significantly higher proportion of granulocytes was detected in 58

the peripheral blood of the infected group after day 7. Additionally, both acute and 59

chronic histological manifestations were observed in the infected group. This model is 60

useful for elucidating the pathophysiology of both acute and chronic osteomyelitis and 61

to assess the effect of novel antibiotics or antibacterial implants. 62


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


4

INTRODUCTION 63

One of the most serious problems in the orthopaedic field is infectious 64

osteomyelitis, which causes progressive inflammation and destruction to bone tissue 65

[18]. Treatment of infectious osteomyelitis is challenging, because the pathogenic 66

organisms and their drug sensitivity are variable. This problem is compounded by 67

increasing numbers of drug-resistant bacterial strains, implant-associated infections, 68

and elderly patients with compromised immune systems [17]. Although progress has 69

been made, controlling infectious osteomyelitis is still difficult. Therefore, 70

experimental studies are warranted to develop more effective treatment options. A 71

number of animal models of osteomyelitis have been reported [13, 25, 33, 35]. 72

However, many of them require sacrifice of the animals, thus they are limited in their 73

availability for real-time assessments of the severity of infection and the efficacy of 74

treatments. As a result, the pathophysiology of osteomyelitis remains poorly 75

understood. 76

A recent development in optical imaging, bioluminescence imaging (BLI), 77

permits the non-invasive sequential monitoring of cell growth and gene expression in 78

vivo. This method allows real time monitoring of implanted cells in live animals [16, 79

26]. Inoculated bacteria that emit a constant bioluminescent signal can be detected 80

through the tissues of a live animal using an ultrasensitive, cooled charged-coupled 81


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


5

device (CCD) camera. This approach has proven useful in studies in the field of 82

oncology [38], endocrine disruptors [27], metabolism [37], hematopoietic cells [41], 83

regenerative medicine [16, 26], immunology [6, 20], and infections [2, 5, 10, 14, 19, 84

23, 34, 42]. 85

Previous models of infectious diseases required sacrifice of the animals to 86

quantify the bacterial numbers by culturing tissue specimens. In contrast, the BLI 87

technique monitors bacterial growth throughout the course of infection in real-time 88

without sacrificing the animals. To the best of our knowledge, there is no previous 89

model, in which infectious processes in bone from the acute to chronic phases were 90

evaluated using BLI, as well as the kinetics of the immune cells and the serum levels 91

of cytokines/chemokines. The purpose of the present study was to establish a real-time, 92

quantitative, and reproducible mouse model of osteomyelitis using the BLI technique. 93

94


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


6

MATERIALS AND METHODS 95

Bioluminescent bacteria 96

A bioluminescent strain of Staphylococcus aureus (S. aureus) Xen-29 was 97

obtained from Caliper LS Co. (Hopkinton, MA) and cultured in Luria Bertani medium 98

(Sigma-Aldrich Co., St. Louis, MO) at 37°C, under ambient aeration with gentle 99

agitation. The bacteria were selectively grown in a medium containing 200 μg/ml 100

kanamycin. S. aureus Xen-29, derived from the parental strain American Type Culture 101

Collection (ATCC) 12600, has a stable copy of a modified Photorhabdus luminescens 102

luxABCDE operon, encoding enzymes responsible for the luminescent reaction. 103

Bacterial bioluminescence requires no substrate to be added to generate the light, 104

thereby constitutively emitting a bioluminescent signal as long as the organism is 105

viable. The bacteria samples were frozen and stored at -80°C. The samples were 106

thawed at 4°C for one hour prior to each experiment. Typically, the bacterial viability 107

was maintained at 4°C for approximately 5 hours after thawing. 108

109

Mouse osteomyelitis model 110

Eighteen BALB/c adult male mice (12-weeks old, 20-25 g) purchased from 111

Sankyo Labo Service (Shizuoka, Japan) were used in this study. Mice were maintained 112


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


7

in our animal facility under specific pathogen-free conditions. The mice were 113

anesthetized with an intraperitoneal injection of 50 mg/kg of pentobarbital, and the 114

skin on their left knee was shaved and sterilized with povidone iodine. A skin incision 115

was made over the left knee, and the distal femur was exposed through a lateral 116

parapatellar arthrotomy with medial displacement of the quadriceps-patellar complex. 117

The distal end of the femur was perforated using a high-speed drill with a 0.5-mm 118

sharp steel burr (Fine Science Tools Inc., Foster city, CA). Then, a channel was created 119

using a 23-G (external diameter, 0.6 mm) needle, through which the bioluminescent 120

strain of S. aureus (1.0x108 CFU) in 1 μl of medium was inoculated into the medullary 121

cavity of the femur using a Hamilton syringe. Phosphate buffered saline (PBS) was 122

administered to the control group using the same technique. The burr hole was closed 123

with bone wax, the dislocated patella was reduced, and the muscle and skin openings 124

were closed by sutures. The animals were placed on a heating pad and carefully 125

monitored until recovery. The observation of spontaneous forelimb movement and the 126

drinking of water were the criteria used to determine that the animals had recovered 127

from the anesthesia. 128

To measure and analyze the bacterial bioluminescent signal by BLI, the mice 129

were anesthetized via inhalation of aerosolized isoflurane mixed with oxygen. The 130

animals were laid on their back and imaged for 5 minutes. All experiments were 131

approved by the Animal Care and Use Committee of Keio University. 132


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


8

Bioluminescent imaging (BLI) 133

A Caliper LS-IVIS® Lumina cooled CCD optical macroscopic imaging system 134

(Summit Pharmaceuticals International Co., Tokyo, Japan) [29] was used for the BLI. 135

Photon emissions of the bacterial bioluminescent signal were captured, converted to 136

false color photon count images, and quantified with Living Image version 3.0 137

software (Caliper LS Co., Hopkinton, MA). The bacterial photon intensity (PI) was 138

expressed as photon flux, in units of photons/sec/cm2/steradian. To quantify the 139

bacterial luminescence, regions of interest (ROIs) were defined in the bacteria plates or 140

inoculated areas and examined. To evaluate the luminescence expression of the 141

bacteria, we first examined whether various numbers of bacteria correlated with the 142

bacterial PI in vitro and in vivo. To analyze the time course of the infection in vivo, the 143

bacterial PI in an ROI was sequentially measured on days 1, 3, 7, 14, and 21 after the 144

operation. 145

146

Serological evaluation 147

Blood samples were collected from the infected and the control mouse groups 148

by retro-orbital bleeding before surgery (day 0) and on days 0.5 (12 hours), 1, 3, 7, 14, 149

and 21 after the operation. To measure inflammatory cytokines and chemokines, the 150


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


9

sera of both groups were serially diluted, and interleukin-6 (IL-6), interleukin-1β 151

(IL-1β), C-reactive protein (CRP), and monocyte chemotactic protein-1 (MCP-1) were 152

measured by using enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems, 153

Minneapolis, MN; Kamiya Biomedical Co., Seattle, WA). Detection was carried out 154

according to the manufacturer’s instructions. 155

156

Flow cytometry 157

Peripheral blood samples from the infected and the control mice were 158

subjected to double immunofluorescence staining and analyzed by flow cytometry on 159

days 1, 3, 7, 14, and 21 after the operation. FITC-anti-CD11b (clone M1/70) and 160

PE-anti-Ly-6C (Gr-1) (clone RB6-8C5) antibodies (Abs) were purchased from BD 161

Biosciences (San Diego, CA). To block the nonspecific binding of Abs to Fc receptors, 162

the isolated cells were incubated with an anti-CD16/32 mAb (clone 2.4G2) (1:250) at 163

4°C for 20 min. The cells were then stained with a mixture of fluorochrome-labeled 164

mAbs at 4°C for 20 minutes, washed, and incubated with 7-Aminoactinomycin D 165

(1:500) (BD Biosciences, San Diego, CA) at 4°C for 5 minutes. Flow cytometry was 166

performed on a FACS Calibur (BD Biosciences, San Diego, CA), and the data were 167

analyzed with FlowJo software (Tree Star, Ashland, OR). Murine granulocytes were 168

defined as SSChighCD11b+ cells [28]. These cells also expressed Gr-1 (not shown). 169


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


10

Histological analysis 170

Femur specimens were collected and analyzed histologically on days 3, 7, 21, 171

and 28 after the operation in both groups. The mice were sacrificed, and the femurs 172

were removed and separated from the soft tissues. The samples were fixed in 4% 173

paraformaldehyde, and demineralized with ethylenediaminetetraacetic acid. The 174

samples were then embedded in paraffin, cut into 5-μm-thick sections, and stained 175

with Hematoxylin and Eosin or Gram. 176

177

Statistical analysis 178

Correlations between the bacterial CFU and the bacterial PI in vitro and in 179

vivo were analyzed by linear regression. Changes in the bacterial PI in the infected 180

group were analyzed with Student’s t-test. One-way ANOVA and the Fisher post hoc 181

test were used to compare the serum levels of IL-6, IL-1β, CRP, and MCP-1, and the 182

proportion of granulocytes in the peripheral blood between the two groups. Correlation 183

between the bacterial PI and the serum CRP levels was determined using Pearson’s 184

correlation coefficient. Dr STSS II software (IBM-SPSS, Tokyo, Japan) was used, and 185

a P-value of less than 0.05 was considered significant in all the statistical analyses. 186

187


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


11

RESULTS 188

Correlation between bacterial number and bacterial photon intensity in vitro 189

A bioluminescent signal that was sufficient to yield a significant value over 190

background was detected by the BLI system from a single colony of the 191

bioluminescent strain of S. aureus Xen-29 cultured in Luria Bertani medium (Figure 192

1A and 1B). To examine the sensitivity of the BLI, we used a CCD-based macroscopic 193

detector to measure the PI of bacterial samples with 7.8x105 to 1.0 x 108 CFU/well. A 194

minimum of 7.8x105 CFU of bacteria was sufficient to produce a detectable signal 195

above the background noise. This quantitative, bioluminescent analysis revealed that 196

there was a significant correlation between the number of bacterial CFU and the 197

bacterial PI in vitro (R2 = 0.998) (Figure 1C). To confirm that only live S. aureus 198

naturally emitted the luminescent signals, colonies of S. aureus fixed with 4% 199

paraformaldehyde were visualized with the BLI system. No signal was detected from 200

the fixed bacteria (data not shown). 201

202

Correlation between bacterial number and bacterial photon intensity in vivo 203

To visualize the infected site ex vivo, immediately after the intra-femur inoculation 204

of S. aureus, the infected femur was removed and separated from the soft tissues, and 205


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


12

the exposed femur was monitored by the BLI system. The bacterial bioluminescent 206

signals were only detected in the medullary cavity of the femur, and not in the 207

surrounding tissue (Figure 2A). To examine whether the number of inoculated bacteria 208

correlated with the bacterial PI in vivo, we performed the inoculation with different 209

numbers of bacteria (1.0 x 108 to 6.0 x 108 CFU per inoculation) and measured the 210

bacterial PI. As shown in Figure 2B, there was a significant correlation between the 211

number of inoculated bacteria and the bacterial PI (R2 = 0.999). 212

213

Time-course of bacterial photon intensity in the mouse OM model 214

Immediately after the inoculation of S. aureus (1.0 x 108 CFU) in 1 μl of medium 215

into the femur, stable luminescent signals were observed in all the animals. Sequential 216

analyses of the bacterial luminescence revealed that the mean bacterial PI in the 217

infected group peaked on day 3 (7.2 ± 1.0x105 PI) and remained at a high level until 218

approximately day 7 (5.2 ± 0.7x105 PI) (Figure 3). Notably, the strong bacterial 219

bioluminescent signal was detected only at the injection site of the femur, and the 220

surrounding tissue was free of infection for 3 months after surgery (data not shown). 221

These observations indicated that this novel mouse model is reproducible and suitable 222

for evaluating the pathophysiology of both acute and chronic osteomyelitis. 223

224


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


13

Serological evaluation 225

During the early phase of infection, the mean serum IL-6 and IL-1β levels in 226

the infected group were elevated. The serum IL-6 level was significantly higher in the 227

infected group than in the control group on day 7 (P< 0.05) (Figure 4A). The mean 228

serum level of IL-1β in the infected group was significantly higher on days 7 and 14 229

(P< 0.05) (Figure 4B). On day 0.5 (12 hours), the mean level of MCP-1 was 230

significantly higher in the infected group than in the control group (P< 0.001) (Figure 231

4C). The mean serum CRP level increased quickly in both groups and remained at 20 232

ng/ml in for 3 days, after which the level remained significantly higher in the infected 233

group, on days 7, 14, and 21 (P< 0.001) (Figure 5A). There appeared to be a direct 234

correlation between the bacterial PI and the serum CRP level in the samples obtained 235

on days 14 and 21 (N = 3 each), the chronic phase of infection (r = 0.85, P< 0.05) 236

(Figure 5B). 237

238

Flow cytometry 239

Flow cytometric analyses using anti-CD11b and Gr1 mAbs showed the 240

presence of granulocytes in the peripheral blood in both groups (Figure 6). The 241

proportion of SSChighCD11b+ granulocytes in the peripheral blood was significantly 242

higher in the infected group than in the control group on days 7, 14, and 21 (N = 4 243


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


14

each) (Figure 6C). 244

245

Histological analysis 246

On day 21, the femur bone marrow from the sham-treated mice contained the 247

normal cellular components of bone marrow: megakaryocytes, erythroid cells, and 248

myeloid cells. In contrast, on day 3, bacterial colonies were detected in the medullary 249

cavity of the infected mouse femur, along with a marked infiltration of neutrophils. 250

The bacterial colonies were Gram-positive. New bone formation started beneath the 251

periosteum on day 7. By day 21, new bone formation and trabecular bone resorption by 252

osteoclasts were present. Manifestations of chronic osteomyelitis such as sequestrum, 253

new bone formation, and fibrosis were prominent on day 28 (Figure 7). 254

255

256

DISCUSSION 257

Osteomyelitis is a serious infectious disease characterized by progressive bone 258

destruction and formation [1, 18]. In most cases, chronic osteomyelitis requires the 259

administration of antibiotic drugs for prolonged periods, and sometimes, surgical 260

procedures. Recently, the incidence of serious nosocomial infection due to 261


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


15

multiple-drug-resistant strains of bacteria has risen. Thus, the treatment of 262

osteomyelitis has become more difficult [32, 44]. Additional sources of rapidly 263

spreading infections include orthopaedic implants, such as those used in fracture 264

fixations, arthroplasty, and spinal surgery [8, 39]. A number of infection models have 265

been created to study the diagnosis and treatment of osteomyelitis. For example, some 266

investigators have attempted to implant staphylococci intravenously or directly into the 267

bone. Although they successfully produced bone infections, these lesions were not 268

progressive enough to simulate human osteomyelitis [31]. Scheman et al. [35] 269

established a reproducible model of chronic osteomyelitis in rabbits by injecting 270

sodium morrhuate and S. aureus directly into the tibial metaphysis. Experimental 271

models using small animals such as rats and mice allow easy handling and are 272

cost-effective; in particular, mice are especially useful for understanding the 273

pathophysiology of osteomyelitis because various genetically modified mice are 274

commercially available. In recent papers, tibia infected mouse models have been used 275

to evaluate implant-associated osteomyelitis [19, 36]. However, because the mouse has 276

tibial curvature with a short medullary cavity and scant surrounding soft tissue, the 277

preparation of this model is technically difficult, often associated with incidental tibial 278

fractures or leakage of the inoculated bacteria. In comparison, our novel osteomyelitis 279

model using mouse femur is easy and reproducible, because the medullary cavity is 280

straight, with a long, thick cortex, and adequate soft tissues surrounding the bone. 281


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


16

In previous animal models of osteomyelitis, the animals had to be sacrificed to 282

quantify the bacterial burden and to assess the extent of infection and inflammation [7]. 283

The experiments using such models are very time consuming, and there is an increased 284

possibility of technical errors during sampling. Furthermore, since the animals have to 285

be sacrificed at certain time points, it is impossible to monitor the same animal 286

throughout the course of the infection. In contrast, recently developed BLI techniques 287

enable us to monitor sequential gene-expression patterns, viabilities of the implanted 288

cells or inoculated bacteria throughout the course of diseases, without sacrificing the 289

animal. Moreover, appropriately prepared animals can be selected at the outset of the 290

experiment, because the bacterial bioluminescence is visible and can be quantified 291

immediately, thus enabling the accurate evaluation of a treatment and avoiding 292

unnecessary follow-ups. Several studies have shown the advantages of in vivo BLI for 293

the real-time monitoring of bacterial infections and their treatment [2, 5, 10, 14, 19, 23, 294

34, 42]. In our mouse osteomyelitis model, sequential analysis of the bacterial 295

luminescence revealed that the bacterial signal peaked on day 3 after the inoculation 296

and then plateaued until day 7, and could be visualized for over 3 months. This time 297

course is similar to that in the previous osteomyelitis models [2, 19], in which the 298

innate immune system contributes to inhibiting the growth of the bacteria at the early 299

phase [19]. Recently, Bernthal et al. [2] established a mouse model of 300

implant-associated infection, as a pre-clinical screening tool. However, a limitation of 301


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


17

their model is the use of the SH1000 S. aureus bioluminescent strain, in which the lux 302

genes are contained in a plasmid. The plasmid is only stable for the first 3 days of 303

broth culture, so it is difficult to estimate bacterial numbers by the bioluminescent 304

signals of this strain after 3 days. Thus, their model represents only acute, not chronic 305

infection. In our model, the lux genes were inserted into the S. aureus chromosome, 306

and the bioluminescent signals were maintained for a longer period. In addition, 307

leakage of the inoculated bacteria to surrounding tissues is often observed in 308

implant-associated osteomyelitis and joint infection models, and can be a major cause 309

of skin ulcer. Such models are poorly reproducible, with short, unstable emission of 310

the bacterial luminescence. In contrast, the pinhole created in the femur for bacterial 311

inoculation in our model closed spontaneously and rapidly enough to keep the bacterial 312

infection contained inside the medullary cavity of the femur. In fact, a strong bacterial 313

bioluminescent signal was detected for over 3 months after the inoculation in our 314

model. 315

In the present study, the mean serum concentrations of IL-6 and IL-1β in the 316

infected group were significantly higher than those in the control group. Marriott et al. 317

[21] demonstrated that osteoblasts express IL-6 during bacterial bone infection in a 318

mouse model and in human bone tissues. IL-6 [11, 12] and IL-1β [4, 24], which are 319

produced by stimulated monocytes/macrophages, stimulate osteoclasts and lead to 320

bone resorption. Yoshii et al. reported that the local levels of IL-6 and IL-1β in the 321


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


18

infected bone were elevated in the early post-infection period in a staphylococcal 322

osteomyelitis model. They suggested that the elevated IL-6 and IL-1β levels induced 323

by infection may be related to bone damage mainly in the early phase of infection [43]. 324

Our results also demonstrated that the mean serum IL-6 and IL-1β concentrations in 325

the infected group were significantly higher at the time pathological changes, such as 326

new bone formation beneath the periosteum, appeared (approximately on day 7). The 327

serum CRP level in both groups was elevated during the first three days (days 0.5, 1, 328

and 3), after which the high level of CRP in the infected group was prolonged through 329

day 7. The elevated CRP level in the early phase might have been caused by the 330

surgical invasion. However, the elevated level was sustained in the infected group for a 331

longer period. The CRP level is one of the most valuable markers for evaluating 332

infectious processes in the clinical field [9, 15, 40]. In this study, there was a high 333

correlation between the CRP level and bacterial PI in vivo during the chronic phase, 334

suggesting that our model is useful for real-time, noninvasive monitoring of the 335

chronic inflammatory processes in osteomyelitis. In contrast, the mean serum level of 336

MCP-1 was significantly higher in the infected group on day 0.5 (12 hours). Cultured 337

osteoblasts produce high MCP-1 levels in response to S. aureus, leading to a proposal 338

that MCP-1 causes the inflammation that results in progressive bone destruction [3]. 339

Marriott et al. [22] reported that increased MCP-1 is the pivotal inflammatory 340

chemokine during S. aureus-associated osteomyelitis in vivo. We also demonstrated 341


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


19

here that the proportion of granulocytes in the peripheral blood was significantly 342

higher in the infected group than in the control group after day 7, suggesting that 343

granulocytes are induced by chemokine activities during the early phase, and that 344

systemic infection due to osteomyelitis was maintained during the chronic phase. Thus, 345

the present osteomyelitis model mimics well the infectious processes in humans. 346

Additionally, the histological study also demonstrated the reproducibility of 347

the present model. The histological analysis showed new bone formation beneath the 348

periosteum at the early phase, and trabecular bone resorption by osteoclasts and 349

fibroblast proliferation during the chronic phase, demonstrating the pathological 350

features of chronic osteomyelitis. 351

In conclusion, we have successfully visualized and quantified the bacterial 352

growth in a mouse osteomyelitis model using in vivo BLI. We were able to monitor the 353

infectious processes throughout the course of the disease in both the acute and chronic 354

phases, without sacrificing the animals. To our knowledge, this is the first report 355

describing a real-time, quantitative, and reproducible model for both acute and chronic 356

osteomyelitis of the mouse femur with kinetics of immune cells and serum levels of 357

cytokines/chemokines. This novel, quantitative, and reproducible model can be used to 358

clarify the pathology and kinetics of osteomyelitis, and evaluate novel in vivo 359

therapeutic strategies including the development of new antibiotics and 360

bacteria-resistant implants, before performing studies in larger animals and human 361


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


20

subjects. 362

363

364

365


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


21

ACKNOWLEDGEMENTS 366

The authors thank Ms Y. Baba for technical assistance with the serological 367

analyses. This work was supported by a Grant from the Japan Orthopaedics and 368

Traumatology Foundation, Inc. No.192, Keio Gijuku Academic Development Funds, 369

The General Insurance Association of Japan, Health Labour Sciences Research Grant 370

and Research for Promoting Technological Seeds. 371


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


22

FIGURE LEGENDS 372

Figure 1: Correlation between bacterial number and bacterial photon intensity in 373

vitro 374

Photon emission of the bacterial bioluminescent signals of Staphylococcus aureus (S. 375

aureus) strain Xen-29 was captured as false color photon count images and quantified 376

by a bioluminescence imaging (BLI) system. To examine the sensitivity of the BLI, a 377

CCD-based macroscopic detector was used to quantify the bacterial photon intensity 378

(PI = photons/sec/cm2/steradian) at various bacterial numbers (7.8 x 105 to 1.0 x 108 379

CFU per well). Bioluminescent signals were detected from colonies of bioluminescent 380

S. aureus in vitro (A and B), and there was a significant correlation between the 381

number of bacterial CFU and the bacterial PI in vitro (R2 = 0.998) (C). 382

Figure 2: Correlation of bacterial number and bacterial photon intensity in vivo 383

During ex vivo imaging, the bacterial bioluminescent signal was detected only in the 384

medullary cavity of the femur, and not in the surrounding tissue (A). Different amounts 385

of bacteria (1.0 x 108 to 6.0 x 108 CFU per inoculation) were inoculated into the femurs, 386

and the bioluminescence in the regions of interest (ROI) was monitored by the BLI 387

system. A significant correlation was observed between the inoculated bacterial 388

number and the bacterial PI in vivo (R2 = 0.999) (B). 389


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


23

390

Figure 3: Time-course changes of bacterial photon intensity in the mouse 391

osteomyelitis model 392

S. aureus strain Xen29 (1.0 x 108 CFU) in 1 μl of medium was inoculated into the 393

medullary cavity of the femur. The bacterial PI from an ROI was then sequentially 394

measured on day 1, 3, 7, 14, and 21 after surgery (N = 6 for each time point). The 395

mean bacterial PI in the infected group peaked on day 3 (7.2 ± 1.0 x 105 PI) and 396

remained at a high level until approximately day 7 (5.2 ± 0.7 x 105 PI). Bars, mean ± 397

SEM. 398

399

Figure 4: Serological evaluation of the control and the infected groups 400

Blood samples from the site of retro-orbital bleeding were collected from the mice 401

before surgery (day 0) and on days 0.5 (12 hours), 1, 3, 7, 14, and 21 after the 402

operation in the infected and the control groups (N = 3, 3). The serum interleukin-6 403

(IL-6) (A), interleukin-1β (IL-1β) (B), and monocyte chemotactic protein-1 (MCP-1) 404

(C) levels were measured with ELISA kits. Bars, mean ± SEM. 405

406

Figure 5: The serum CRP level. Correlation of bacterial Photon intensity and the 407

serum CRP level 408


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


24

The serum C-reactive protein (CRP) (A) levels in the infected and the control groups 409

were also measured with ELISA kits (N = 3, 3). Bars, mean ± SEM. The bacterial CFU 410

and bacterial PI were correlated. The serum CRP concentration was examined in the 411

same samples from mice on days 14 (N = 3) and 21 (N = 3). A direct correlation 412

between the bacterial PI and the serum CRP concentration was observed on both days 413

(r = 0.85, P < 0.05) (B). 414

415

Figure 6: Proportion of granulocytes in the peripheral blood 416

Peripheral blood samples were analyzed by flow cytometry on days 0.5 (12 hours), 1, 3, 417

7, 14, and 21 after the operation in the infected and the control groups. Flow 418

cytometric analyses of SSChighCD11b+ granulocytes in the peripheral blood of the 419

control (A) and the infected (B) mice on day 21 are shown. The proportions of 420

SSChighCD11b+ granulocytes in the peripheral blood on days 7, 14, and 21 were 421

significantly higher in the infected group than in the control group (N = 4, 4) (C). Bars, 422

mean ± SEM. 423

424

Figure 7: Changes over time in the histology of the femurs from the infected and 425

control mice 426

Hematoxylin and eosin staining of longitudinal sections of the non-infected and 427

infected femurs on day 3, 7, 21, and 28 after bacterial inoculation. The middle and 428


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


25

right panels show higher-power views of the white-squared areas of the left panels and 429

black-squared areas of the middle panels, respectively. Inset of the infected femur (day 430

3) indicates Gram-stain-positive bacteria. *, necrotic area with bacterial colonies; 431

arrows, osteoclasts; **, sequestrum. Both acute and chronic manifestations of 432

osteomyelitis were observed in this model. Scale Bars = 100 μm. 433

434

435


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


26

REFERENCES 436

1. Aliabadi, P., and N. Nikpoor. 1994. Imaging osteomyelitis. Arthritis Rheum. 437

37:617-622. 438

2. Bernthal, N. M., A. I. Stavrakis, F. Billi, J. S. Cho, T. J. Kremen, S. I. Simon, 439

A. L. Cheung, G. A. Finerman, J. R. Lieberman, J. S. Adams, and L. S. Miller. 440

2010. A mouse model of post-arthroplasty Staphylococcus aureus joint infection to 441

evaluate in vivo the efficacy of antimicrobial implant coatings. PLoS One. 442

5:e12580. 443

3. Bost, K. L., J. L. Bento, C. C. Petty, L. W. Schrum, M. C. Hudson, and I. 444

Marriott. 2001. Monocyte chemoattractant protein-1 expression by osteoblasts 445

following infection with Staphylococcus aureus or Salmonella. J Interferon 446

Cytokine Res.21:297– 304. 447

4. Boyce, B. F., T. B. Aufdemorte, I. R. Garrett, A. J. P. Yates, and G. R. Mundy. 448

1989. Effects of interleukin-1 on bone turnover in normal mice. Endocrinology. 449

125:1142-50. 450

5. Contag, C. H., P. R. Contag, I. Mullins, S. D. Spilman, D. K. Stevenson, and D. 451

A. Benaron. 1995. Photonic detection of bacterial pathogens in living hosts. Mol. 452

Microbiol. 18:593–603. 453

6. Costa, G. L., M. R. Sandora, A. Nakajima, E. V. Nguyen, C. Taylor-Edwards, 454


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


27

A. J. Slavin, C. H. Contag, C. G. Fathman, and J. M. Benson. 2001. Adoptive 455

immunotherapy of experimental autoimmune encephalomyelitis via T cell delivery 456

of the IL-12 p40 subunit. J Immunol. 167:2379-87. 457

7. Cremieux, A. C., and C. Carbon. 1997. Experimental models of bone and 458

prosthetic joint infections. Clin Infect Dis. 25:1295-302. 459

8. Darouiche, R. O. 2004. Treatment of infections associated with surgical implants. 460

N Engl J Med. 350:1422-9. 461

9. Du Clos, T. W. 2000. Function of C-reactive protein. Ann. Med. 32:274-8. 462

10. Francis, K. P., D. Joh, C. Bellinger-Kawahara, M. J. Hawkinson, T. F. 463

Purchio, and P. R. Contag. 2000. Monitoring bioluminescent Staphylococcus 464

aureus infections in live mice using a novel luxABCDE construct. Infect. Immun. 465

68:3594–3600. 466

11. Ishimi, Y., C. Miyaura, C. H. Jin, T. Akatsu, E. Abe, Y. Nakamura, A. 467

Yamaguchi, S. Yoshiki, T. Matsuda, T. Hirano, T. Kishimoto, and T. Suda. 468

1990. IL-6 is produced by osteoblasts and induces bone resorption. J Immunol. 469

145:3297-3303. 470

12. Kurihara, N., D. Bertolini, T. Suda, Y. Akiyama, and G. D. Roodman. 1990. 471

IL-6 stimulates osteoclast-like multinucleated cell formation in long term human 472

marrow cultures by inducing IL-1 release. J Immunol 144:4226-30. 473

13. Johansson, A., O. Svensson, G. Blomgren, G. Eliasson, and C. E. Nord. 1991. 474


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


28

Anaerobie osteomyelitis. A new experimental rabbit model. Clin Orthop. 475

265:297–301. 476

14. Kadurugamuwa, J. L., L. V. Sin, J. Yu, K. P. Francis, R. Kimura, T. F. Purchio, 477

and P. R. Contag. 2003. Rapid direct method for monitoring antibiotics in a 478

mouse model of bacterial biofilm infection. Antimicrob Agents Chemother. 479

47:3130-7. 480

15. Khan, M. H., P. N. Smith, N. Rao, and W. F. Donaldson. 2006. Serum C-reactive 481

protein levels correlate with clinical response in patients treated with antibiotics for 482

wound infections after spinal surgery. Spine J. 6:311-5. 483

16. Kim, D. E., D. Schellingerhout, K. Ishii, K. Shah, and R. Weissleder. 2004. 484

Imaging of stem cell recruitment to ischemic infarcts in a murine model. Stroke. 485

35:952-7.2. 486

17. Kurtz, S. M., K. L. Ong, J. Schmier, F. Mowat, K. Saleh, E. Dybvik, J. 487

Kärrholm, G. Garellick, L. I. Havelin, O. Furnes, H. Malchau, and E. Lau. 488

2007. Future clinical and economic impact of revision total hip and knee 489

arthroplasty. J Bone Joint Surg Am. 89:144–51. 490

18. Lew, D. P., and F. A. Waldvogel. 2004. Osteomyelitis. Lancet. 364:369-79. 491

19. Li, D., K. Gromov, K. Soballe, J. E. Puzas, R. J. O'Keefe, H. Awad, H. Drissi, 492

and E. M. Schwarz. 2008. Quantitative mouse model of implant-associated 493

osteomyelitis and the kinetics of microbial growth, osteolysis, and humoral 494


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


29

immunity. J Orthop Res. 26:96-105. 495

20. Mandl, S., C. Schimmelpfennig, M. Edinger, R. S. Negrin, and C. H. Contag. 496

2002. Understanding immune cell trafficking patterns via in vivo bioluminescence 497

imaging. J Cell Biochem Suppl. 39:239-48. 498

21. Marriott, I., D. L. Gray, S. L. Tranguch, V. G. Fowler Jr, M. Stryjewski, L. 499

Scott Levin, M. C. Hudson, and K. L. Bost. 2004. Osteoblasts express the 500

inflammatory cytokine interleukin-6 in a murine model of staphylococcus aureus 501

osteomyelitis and infected human bone tissue. Am J Pathol. 164:1399-406. 502

22. Marriott, I., D.L. Gray, D.M. Rati, V.G. Fowler Jr., M.E. Stryjewski, L.S. 503

Levin, M.C. Hudson, and K.L. Bost. 2005. Osteoblasts produce monocyte 504

chemoattractant protein-1 in a murine model of Staphylococcus aureus 505

osteomyelitis and infected human bone tissue. Bone. 37:504-12. 506

23. Mortin, L. I., T. Li, A. D. Van Praagh, S. Zhang, X. X. Zhang, and J. D. Alder. 507

2007. Rapid bactericidal activity of daptomycin against methicillin-resistant and 508

methicillin-susceptible staphylococcus aureus peritonitis in mice as measured with 509

bioluminescent bacteria. Antimicrob Agents Chemother. 51:1787-94. 510

24. Nishihara, T., Y. Ishihara, T. Noguchi, and T. Koga. 1989. Membrane IL-1 511

induces bone resorption in organ culture. J Immunol. 143:1881-6.) 512

25. Norden, C. W. 1970. Experimental osteomyelitis. I. A description of the model. J 513

Infect Dis. 122:410–8. 514


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


30

26. Okada, S., K. Ishii, J. Yamane, A. Iwanami, T. Ikegami, H. Katoh, Y. Iwamoto, 515

M. Nakamura, H. Miyoshi, H. J. Okano, C. H. Contag, Y. Toyama, and H. 516

Okano. 2005. In vivo imaging of engrafted neural stem cells: its application in 517

evaluating the optimal timing of transplantation for spinal cord injury. FASEB J. 518

19:1839-41. 519

27. Pillon, A., N. Servant, F. Vignon, P. Balaguer, and J. C. Nicolas. 2005. In vivo 520

bioluminescence imaging to evaluate estrogenic activities of endocrine disrupters. 521

Anal Biochem. 340:295-302. 522

28. Pucci, F., M. A. Venneri, D. Biziato, A. Nonis, D. Moi, A. Sica, C. Di Serio, L. 523

Naldini, and M. De Palma. 2009. A distinguishing gene signature shared by 524

tumor-infiltrating Tie2-expressing monocytes, blood "resident" monocytes, and 525

embryonic macrophages suggests common functions and developmental 526

relationships. Blood. 114:901-14. 527

29. Rice, B. W., M. D. Cable, and M. B. Nelson. 2001. In vivo imaging of 528

light-emitting probes. J Biomed Opt. 6:432-40. 529

30. Richard, A. D., H. C. Jennifer, L. B. Caroline, D. Alain, E. S. Amanda, and M. 530

O. Christopher. 2008. Macrophage-specific metalloelastase (MMP-12) truncates 531

and inactivates ELR+CXC chemokines and generates CCL2, -7, -8, and -13 532

antagonists: potential role of the macrophage in terminating polymorphonuclear 533


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


31

leukocyte influx. Blood. 112: 3455-64. 534

31. Rissing, J. P. 1990. Animal models of osteomyelitis. Knowledge, hypothesis, and 535

speculation. Infect Dis Clin North Am. 4:377-90. 536

32. Rissing, J. P. 1997. Antimicrobial therapy for chronic osteomyelitis in adults: role 537

of the quinolones. Clin Infect Dis. 25:1327-33. 538

33. Rissing, J. P., T. B. Buxton, R. S. Weinstein, and R. K. Shockley. 1985. Model 539

of experimental chronic osteomyelitis in rats. Infect Immun. 47:581–6. 540

34. Sadikot, R. T., E. D. Jansen, T. R. Blackwell, O. Zoia, F. Yull, J. W. 541

Christman, and T. S. Blackwell. 2001. High-dose dexamethasone accentuates 542

nuclear factor-kappa b activation in endotoxin-treated mice. Am J Respir Crit Care 543

Med. 164:873–8. 544

35. Scheman, L., M. Janota, and P. Lewin. 1941. The production of experimental 545

osteomyelitis: preliminary report. JAMA. 177:1525-29. 546

36. Schmidmaier, G., M. Lucke, B. Wildemann, N. P. Haas, and M. Raschke. 2006. 547

Prophylaxis and treatment of implant-related infections by antibiotic-coated 548

implants: a review. Injury. 37:S105-12. 549

37. Sever, D., R. Eldor, G. Sadoun, L. Amior, D. Dubois, C. Boitard, C. Aflalo, and 550

D. Melloul. 2011. Evaluation of impaired beta-cell function in nonobese-diabetic 551

(NOD) mouse model using bioluminescence imaging. FASEB J. 25:676-84. 552

38. Sweeney, T. J., V. Mailänder, A. A. Tucker, A. B. Olomu, W. Zhang, Y. Cao, R. 553


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


32

S. Negrin, and C. H. Contag. 1999. Visualizing the kinetics of tumor-cell 554

clearance in living animals. Proc Natl Acad Sci U S A 96:12044-9. 555

39. Toms, A. D., D. Davidson, B. A. Masri, and C. P. Duncan. 2006. The 556

management of peri-prosthetic infection in total joint arthroplasty. J Bone Joint 557

Surg Br. 88:149-55. 558

40. Volanakis, J. E. 2001. Human C-reactive protein: expression, structure, and 559

function. Mol. Immunol. 38:189-97. 560

41. Wang, X., M. Rosol, S. Ge, D. Peterson, G. McNamara, H. Pollack, D. B. Kohn, 561

M. D. Nelson, and G. M. Crooks. 2003. Dynamic tracking of human 562

hematopoietic stem cell engraftment using in vivo bioluminescence imaging. 563

Blood. 102:3478-82. 564

42. Wiles, S., K. M. Pickard, K. Peng, T. T. MacDonald, and G. Frankel. 2006. In 565

vivo bioluminescence imaging of the murine pathogen citrobacter rodentium. 566

Infect Immun. 74:5391-6. 567

43. Yoshii, T., S. Magara, D. Miyai, H. Nishimura, E. Kuroki, S. Furudoi, T. 568

Komori, and C. Ohbayashi. 2002. Local levels of interleukin-1β, -4, -6, and 569

tumor necrosis factor in an experimental model of murine osteomyelitis due to 570

staphylococcus aureus. Cytokine. 19:59-65. 571

44. Zinner, S. H. 1999. Changing epidemiology of infections in patients with 572

neutropenia and cancer: emphasis on gram-positive and resistant bacteria. Clin 573


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/


33

Infect Dis. 29:490-4. 574

575

576

577


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Figure 1

Phase

A B

BLI

2 48836

C5.0

y = 2073.7e2.4883x

R² = 0.9982(×106 photons

/sec/cm2 )

coun

ts

2.0

3.0

4.0

0.5 1.5 2.5 3.5

Phot

on

1.0

(×107 CFU log10)

Bacterial number


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Figure 2

Ain vivo

ex vivo

(×105 photons

B

y = 1.2428x - 0.014R² = 0.9993

4

6

8

4.0

6.0

8.0( 10 photons

/sec/cm2 )

oton

cou

nts

0

2

0 2 4 6 8

2.0

2.0 4.0 6.0(×108 CFU )

8.0

Bacteria number

Pho


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Figure 3

A

day21day1 day3 day7 day14

***10.0

B(×105 photons

/sec/cm2 )

ton

coun

ts

***

*

4 0

6.0

8.0

1 3 7 14 21 (Days)

Phot

2.0

4.0

(Days)

* P<0.05*** P<0.001


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

(pg/ml)A B

Figure 4

( / l)

500

600

700

800

L-6

(pg/ml)

*

20

25

30

35

β

(pg/ml)

*

*

0

100

200

300

400

0 0 1 3 14

IL

(Days)0

5

10

15

0 0 5 1 3 7 14 (Days)IL

-10 0.5 1 3 7 14

C

( y ) 0 0.5 1 3 7 14 (Days)

(pg/ml)

* P<0.05*** P<0.001infected

control

40

50

60

70

80

MC

P-1

***

0

10

20

30

0 0.5 1 3 7 14 21

M

(Days)


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

A

Figure 5

(ng/ml)

20

25

30(ng/ml)

***

******

P

0

5

10

15

CR

P

00 0.5 1 3 7 14 21

B

(Days)

*** P<0.001infectedcontrol

17

17.5( ng/ml )

r= 0.85, P<0.05

B

15

15.5

16

16.5

CR

P

14.5

15

100000200000300000400000500000600000700000

Photon counts

2.0 3.0 4.0 5.0 6.0 7.01.0(×105 photons

/sec/cm2 )


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Figure 6

A BControl (day 21) Infected (day 21)( y ) ( y )

70

80

C***

Ratio of granulocyte

30

40

50

60

70 ******

0

10

20

1 3 7 14 21 (Days)

*** P<0.001infectedcontrol


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Figure 7

Sham (day 21)

OM (day 3)*

OM (day 7)

OM (day 21)

**OM (day 28)


http://iai.asm.org/

Dow

nloaded from

http://iai.asm.org/

Documents

Downloaded from on January 9, 2020 by ... · Establishment of mouse osteomyelitis model 4 63 INTRODUCTION 64 One of the most serious problems in the orthopaedic field is infectious