A new multiple trauma model of the mouse

Experiments were carried out in accordance with the German Animal Welfare Legislation and were approved by the local institutional animal care and research advisory committee and permitted by the local government of Lower Saxony, Germany (AZ 10AO29). The study was performed at the experimental trauma surgery laboratory of Hannover Medical School (MHH).

Experiments were conducted in an operating room at the animal research facility. One hundred twelve male C57BL/6 N mice (Charles River, Germany) weighing 22 ± 3 g (gram) were used for the study. Twenty male mice were used in preliminary experiments to determine the weight needed for induction of chest trauma. All mice were handled at room temperature for 14 days before treatment, and all mice were age-matched (12 weeks old). We used only male C57BL/6 N mice for this primary study because gender of mice affects hormones and cell-mediated immune response [20, 21]. Cytokine expression also differs between male and female mice [22]. Further studies with female mice will be necessary. Animals were maintained under standardized conditions in a controlled environment at 21 ± 2 °C (Celsius), with a relative humidity of 50% and artificial light (14 h light, 10 h dark). They received a commercial pellet diet (altromin 1320, Altromin, Lage, Germany) and water ad libitum. Analgesic treatment was administered to all animals in the form of metamizol-sodium (200 mg/kg (milligram/kg) body weight; Novalgin® Hoechst, Unterschleiβheim, Germany) throughout the study. Mice were injected subcutaneously under deep anesthesia prior to induction of the thorax trauma and after induction of the femur fracture. For postoperative analgesia, 0.8 mg/mL (milligram/ml) Novaminsulfon Lichtenstein 500 mg (Zentiva Pharma GmbH, Frankfurt am Main, Germany) was added to the drinking water for the first 3 days after trauma.

All surgical procedures were performed under deep anesthesia with isoflurane ((Minrad, Bethlehem, PA)) and local application of xylazine (16 mg/kg) (Rompun®, Bayer, Leverkusen, Germany). The mice were warmed to 36 °C using infrared heat lamps after the surgical procedures were complete. Wound closure was performed before recovery from the anesthesia.

For quantification of activity as a measure of clinical status after trauma, a scoring system was used [23].

Animals were randomly assigned to one of three groups. In the first group, mice received an isolated femur shaft fracture after stabilization with a pin (Fx). In the second group, mice received a combined intramedullary femoral fracture stabilization and blunt thoracic trauma (group TTFx). In the third group, mice underwent a control operation with intramedullary pin implantation in an intact femur without fracture (control).

One hundred twelve multiple trauma, femur fracture and control mice were tested (48 multiple trauma and 48 isolated femur fracture). The control group consisted of 16 mice (16,6% (percent)) that underwent an operation (femur stabilization) in the absence of fracture or chest trauma (Fig. 1c). Experiments were undertaken by three different surgeons. There were no significant differences with regard to moribund animals (surgeon 1: 6 moribund mice, surgeon 2: 4 moribund mice, surgeon 3: 6 moribund mice).

After induction of the femur fracture, a blunt thoracic trauma was induced by a modified version of a previously described model for rats of bilateral lung contusion in the TTFx group [13, 24]. The method has been previously described as reproducible.

Blunt thoracic trauma was induced in anesthetized mice by dropping a hollow aluminum cylindrical weight (300 g) from a height of 55 cm (centimeter) through a vertical stainless steel tube onto a Lexon platform resting on the chest (Fig. 1 a, b). The impact energy E (1.617 J (Joule)) of the falling weight was calculated using the eq. E = m x g x h, where m = mass of aluminum weight (in kilograms), g = gravitational acceleration (9.8 ms⁻² (milliseconds⁻²)) and h = height of weight above the Lexon platform (in meters). The calculations assumed that all the potential energy of the weight was transferred to the animal, neglecting frictional dissipation. The platform was suspended on Teflon guides to minimize friction and facilitate energy transfer to the anesthetized animal. The shield was reproducibly placed entirely over the chest without intrusion onto the abdomen.

The experiments were performed by three different autonomous surgeons. All data were examined by a statistician. The cause of death was determined during organ removal immediately after death.

The experimental design of the multiple trauma model is based on a two-hit model. The first hit consists of a closed femur fracture on the right side as described by Bonnarens [14]. In brief, under deep anesthesia with isoflurane, a 20 gauge needle was first inserted into the canal of the mouse femur as an intramedullary pin (Fig. 1 c). After primary wound closure, a standardized femur fracture was induced in both groups using a blunt guillotine device weighing 500 g (0.784 J) after primary stabilization. This procedure resulted in an A-type femoral fracture combined with a moderate soft-tissue injury. The type of fracture (A fracture, AO classification) was controlled after sacrifice.

The body weight, body temperature and activity of mice were measured in all groups before trauma and after trauma before sacrifice.

Samples for the control group were collected using the retrobulbar technique during the preliminary test. Posttraumatic control was performed by heart puncture. By using 2-mL (milliliter) syringes (Pico50, Radiometer Medical, Brønshøj, Denmark) containing 80 IU electrolyte-balanced heparin, blood samples (0.7 ml) for blood gas analysis and assessment of marker enzyme activities were collected from heart. The animals were sacrificed by heart puncture under deep anesthesia. The hemoglobin concentration, hematocrit and metabolic parameters (lactate, glucose), and osmolality were assessed using a blood gas analyzer (ABL 715, Radiometer, Copenhagen, Denmark).

Animals were sacrificed immediately after trauma, after 6 h, 12 h, 24 h, 3 d and 7 d to obtain samples for histologic examination. Tissue samples from lung were collected and stored at −20 °C until processed.

Tissue samples were embedded in paraffin. Sections (5 μm (micrometer)) were obtained from the central portion of the lung with a sliding microtome (HM 430; Microm International), placed on Superfrost Plus microscope slides (Thermo Scientific) and incubated overnight at 60 °C. The sections were routinely stained with hematoxylin and eosin (H&E). Safranin O staining was carried out for 6 min using a 0.1% aqueous solution at pH 3.0.

Chest trauma of multiple trauma mice was assessed by micro computed tomography (μCT).

The CT scan was performed at the Molecular Imaging North Competence Center (Am Botanischen Garten 14, 24,118 Kiel). Micro computed tomography in Kiel has been applied previously to mice in several studies [25, 26]. Two radiologists planed every scan.

The total scan time was approximately 14 min. Scanning of mice lungs has been described previously [27]. The lungs of mice, which were killed immediately after trauma, were scanned using a Novotec MicroScope (Novotec Medical GmbH, Pforzheim) at an isotropic nominal spatial resolution (voxel size) of 15–20 μm. Samples were transported on ice before the scanned lungs were positioned on a special platform to prevent artifacts. Image analysis was performed using ImageJ software.

To analyze concentrations of different cytokines, blood samples obtained by heart puncture of the mice were centrifuged for five minutes. The supernatant was removed and stored at 20 °C until processing. The concentrations of IL-1β, IL-6, IL-10, IL-12p70 and TNFα in plasma samples were analyzed using a Luminex assay according to standard protocols with LiquiChip200 (Qiagen). A Milliplex cytokine multiplex immunoassay kit (MPXHCYTO-60 K-01; Millipore) was used for protein detection.

Statistical analysis was performed using a standard software application (SPSS Inc., Chicago, IL, USA). Differences between the sham group and the other groups were evaluated using the Wilcoxon signed-rank test. For compromise of mice with an isolated fracture and mice with chest trauma and a femur fracture we used further the Mann-Whitney U test. Probability values less than 0.05 were considered statistically significant. The Data are shown as box-and-whisker plot with median and interquartile range.

Regarding the reproducibility of the model in the group of multiple trauma mice, 16 mice (33%) died after chest trauma because of hemorrhage and 32 mice (66%) survived.

All mice that underwent chest trauma and femur fracture showed a tendency of reduced weight after 6 h. Representative the weight loss was shown for 5 mice with an initial weight of 25,62 g, 25,58 g, 26,54 g, 24,40 g and 24,83 g. After 6 h the weight was reduced to 24,06 g, 24,95 g, 25,39 g, 24,16 g and 23,17 g (Fig. 2). After 3 days the weight had returned completely to baseline values. Temperature declined until 24 h after trauma (Fig. 3). Representative temperature was shown for 5 mice before (38,7 °C, 37,6 °C, 37,8 °C, 37,2 °C, 38,7 °C) and 24 h after the trauma (37,2 °C, 36,8 °C, 37,6 °C, 37 °C, 37,5 °C). After 24 h weight and temperature returned to baseline values. There was no significant decrease in either temperature or weight after trauma.

In a previous examination 5 mice of the same age and weight were punctured retrobulbar before trauma induction as a control to measure the baseline hemoglobin values. Compared to the control group (14,8 g/dl (gram/deciliter), 15,3 g/dl, 14,4 g/dl, 14 g/dl, 15,3 g/dl, 14,9 g/dl), hemoglobin declined after 6 h (13,7 g/dl, 13 g/dl, 13,5 g/dl, 11,7 g/dl, 13,7 g/dl, 12,2 g/dl) and 12 h (13,7 g/dl, 11,7 g/dl, 13,2 g/dl, 14,7 g/dl, 12,5 g/dl, 14,1 g/dl) until 24 h after induction of chest trauma (14,4 g/dl, 13,4 g/dl, 13,2 g/dl, 12 g/dl, 12,3 g/dl, 13,7 g/dl) (Fig. 4). Hemoglobin values had returned to baseline values 3 days after trauma. All tested mice showed comparable results.

Macroscopic and microscopic analyses showed that immediate death was caused by intrathoracic bleeding or heart contusion. CT scans were performed on mice that were killed immediately after chest trauma (Fig. 5). Hemothorax and lung contusion could be observed on the thoracic CT. The injured mice had no rib fractures. All mice showed comparable results on CT scan.

Histological sections were examined to assess the severity of pulmonary tissue injury in mice (Fig. 6). HE staining was performed for lung samples of mice with chest trauma and an unoperated control group. At 24 h post-contusion, HE staining of lung samples showed thickening of the alveolar lining with ongoing leukocytic infiltration. All of the stained lungs showed comparable results.

Different cytokines were analyzed by multiplex immunoassay in plasma samples of mice with an isolated femur fracture and in mice with a chest trauma and femur fracture (Fig. 7 a, b, c, d, e). Mice with an isolated femur fracture showed a down-regulation of IL1β from 0 h (808,84 ± 190,93 pg/ml (picogram/ml)) to 12 h (492,81 ± 190,93 pg/ml) and an up-regulation of IL-6 from 0 h (39,32 ± 18,58) to 6 h (76,67 ± 16,40). IL1β and IL-6 were up-regulated from 0 h (IL1β 209,12 ± 166,51 pg/ml; IL-6 22,98 ± 9,59 pg/ml) to 6 h after multiple trauma (IL1β 829,40 ± 163,87 pg/ml; IL-6 99,88 ± 65,18 pg/ml). IL-10 expression was down-regulated from 0 h (2464,06 ± 894,49) to 12 h (1773,15 ± 742,4) in mice with an isolated fracture. In contrast, the maximum of IL-10 expression was reached at 12 h after multiple trauma (2519,12 ± 1782,87 pg/ml), whereas IL-10 expression was reduced directly after trauma (0 h 1085,22 ± 702,85 pg/ml) compared with the control group (1644,08 ± 1001,46 pg/ml).

After an isolated femur fracture, an up-regulation of IL-12p70 could be detected from 0 h (679,24 ± 578,52) to 6 h (1523,32 ± 480,26). Similar to mice with an isolated fracture, IL-12p70 was up-regulated and could be detected from 0 h (321,64 ± 294,74 pg/ml) to 6 h after multiple trauma (1671,78 ± 350,74 pg/ml).

A slight regulation of TNFα expression was observed after an isolated fracture, whereas TNFα expression declined immediately after multiple trauma (0 h 578,93 ± 232,48 pg/ml) compared with the control group (1241,5 ± 266,22 pg/ml). The baseline level of TNFα expression was recovered at 6 h after trauma (1288,76 ± 693,74 pg/ml). After multiple trauma all cytokines (IL1β, IL-6, IL-12p70, IL-10 and TNFα) were up-regulated (Fig. 7 a, b, c, d, e). Mice with multiple trauma showed a significant up-regulation for IL-6 compared to mice with an isolated fracture.