Photothrombosis is a minimally invasive and highly reproducible procedure to induce focal ischemia in the spinal cord and serves as a model of spinal cord injury in mice.
Spinal cord injury (SCI) is a devastating clinical condition causing permanent changes in sensorimotor and autonomic functions of the spinal cord (SC) below the site of injury. The secondary ischemia that develops following the initial mechanical insult is a serious complication of the SCI and severely impairs the function and viability of surviving neuronal and non-neuronal cells in the SC. In addition, ischemia is also responsible for the growth of lesion during chronic phase of injury and interferes with the cellular repair and healing processes. Thus there is a need to develop a spinal cord ischemia model for studying the mechanisms of ischemia-induced pathology. Focal ischemia induced by photothrombosis (PT) is a minimally invasive and very well established procedure used to investigate the pathology of ischemia-induced cell death in the brain. Here, we describe the use of PT to induce an ischemic lesion in the spinal cord of mice. Following retro-orbital sinus injection of Rose Bengal, the posterior spinal vein and other capillaries on the dorsal surface of SC were irradiated with a green light resulting in the formation of a thrombus and thus ischemia in the affected region. Results from histology and immunochemistry studies show that PT-induced ischemia caused spinal cord infarction, loss of neurons and reactive gliosis. Using this technique a highly reproducible and relatively easy model of SCI in mice can be achieved that would serve the purpose of scientific investigations into the mechanisms of ischemia induced cell death as well as the efficacy of neuroprotective drugs. This model will also allow exploration of the pathological changes that occur following SCI in live mice like axonal degeneration and regeneration, neuronal and astrocytic Ca2+ signaling using two-photon microscopy.
Traumatic spinal cord injury (SCI) is a devastating clinical condition affecting the sensorimotor and autonomic functions of the SC. Patients surviving SCI are often left with debilitating paraplegia that significantly affects their daily activities and quality of life1. Experimental SCI models have been an indispensable tool in the scientific investigation to understand the pathophysiology of SCI and associated neural repair processes. These models have also been used to test the preclinical efficacy of various experimental neuroprotective interventions that are aimed at functional recovery. Currently, majority of the SCI models in practice employ the use of physical blunt force to mechanically disrupt and injure the SC. These methods include contusion, compression, dislocation and transection of the SC2. It has been suggested that after the primary mechanical insult a secondary injury in the form of ischemia sets in in the injured SC3,4. The etiology of secondary ischemia includes extensive tissue degeneration, parenchymal hemorrhage and sometimes by obstruction of blood vessels by tissue edema5-7. As a result of the secondary injury the integrity of SC is further affected, neurons and glial cells are severely impaired in function and viability and undergo apoptosis which leads to infarct growth during the chronic stage of injury, analogous to the growth of ischemic penumbra following stroke8,9. Several mechanisms like excitotoxicity, free radical production, and inflammation have been reported to be responsible for the ischemic cell death following SCI10,11. In addition, SC ischemia is a serious complication of thoraco-abdominal aortic aneurysm repair surgeries which often lead to paraplegia in the patients12,13. In spite of such high clinical impact very few models of spinal cord ischemia with high reproducibility are currently available.
Photothrombosis (PT) is a commonly used method for the induction of focal ischemia in the brain14-20. The technique is fairly non-invasive, highly reproducible and produces a precise focal ischemic lesion in the exposed area of the brain17-21. This is achieved by systemic administration of photoactive dyes like Rose Bengal (RB)16-20,22 or erythrosine B23 followed by localized irradiation of blood vessels with proper light source. Photoactivation of the dye causes the generation of free radicals which disrupt the integrity of the smooth vascular endothelium, and cause the platelets to accumulate, which subsequently forms a thrombus. The obstruction of blood flow by the thrombus results in an infarct in the region supplied by the vessel24. Due to ease of control on the intensity and duration of irradiation this procedure yields a highly uniform and reproducible infarct. Furthermore, this method can be employed to induce an infarct at various anatomical locations enabling spatial (e.g., grey matter vs. white matter) understanding of the effect of ischemia.
The aim of the current study is to develop an easy and highly reproducible model of SC ischemia in mice. We described the procedure of a PT model of SC ischemia in mice. Results from histology and immunostaining demonstrated that PT can effectively induce SC infarction, neuronal loss and reactive gliosis.
Note: Mice (C57BL/6J, male) aged 10 – 12 weeks were used in this study. All the procedures were performed in accordance with the NIH Guide for the Care and Use of Laboratory Animals and were approved by the University of Missouri Institutional Animal Care and Use Committee (IACUC).
1. Pre-Surgery
2. Surgical Procedure
3. Induction of PT
4. Post-surgery Care
5. Transcardial Perfusion, Nissl staining and Immunostaining
The aim of this study was to produce spinal cord ischemia in mice using a PT model. After the desired region of bone above the spinal cord (T10 – T12) was thinned, Rose Bengal was injected through retro-orbital sinus route, and ischemia was induced by PT. Figure 1A, B show the mouse positioned in a custom-made surgical platform during surgery. The mouse was held in place by a snout clamp and two adjustable vertebrate clamps to stabilize the spinal cord. Figure 1C show a thinned window above the spinal cord of T10 – T12. The main blood vessel and its branches can be clearly visualized. To confirm the induction of ischemia, changes in blood flow was measured using a laser Doppler flowmeter before and after PT (Figure 2A, B). For analysis, the decrease in % of blood flow was calculated using the baseline blood flow before photothrombosis. Blood flow dropped to ~20% immediately after light illumination as compared with the basal level prior to illumination. Figure 3B, C shows fluorescent images of spinal cord blood vessels at the beginning and the end of PT. Illumination for 2 min induced blood clot in the blood vessels (Figure 3C), suggesting the induction of ischemia, consistent with the measurements of the laser Doppler flowmeter. To inspect the injury caused by PT, mice were sacrificed 3 days after PT and Nissl staining was performed. Images taken following Nissl staining showed the infarct region that can be clearly demarcated from the surrounding region, indicating spinal cord tissue damage and cell death after PT (Figure 4). Immunostaining was performed for NeuN, GFAP and Iba1. NeuN+ neurons were lost in the grey matter in the ischemic core (Figure 5A), while GFAP expression was increased in the border of ischemic core (Figure 5B, also see the boxed region). The Iba1+ microglia exhibited a globoid morphology (i.e., an enlarged cell body with shorter and fewer processes, see the boxed region) along with increased Iba1 expression (Figure 5C). Although there was a tissue loss in the ischemic core region due to floating section staining, an increase in GFAP and Iba1 expression in the entire peri-infarct region can be clearly observed. These results indicate neuronal death and reactive gliosis in the penumbra after SC ischemia. On the other hand, substantial functional deficits were observed in the injured mice, i.e., disabled hind-limb movement one day after PT, indicating paralysis of hind-limbs (see the Movie).
Figure 1. PT-induced ischemia model in spinal cord. (A) Photographs of the surgery platform for the spinal cord PT. Inset: enlarged vertebral clamps. (B) The mouse was held by a snout clamp and by two custom-made vertebral clamps on the stage. Notice that the bone was thinned at T10 – T12 region and two metal vertebral clamps were used to stabilize the spinal cord. (C) A zoom-in image showing the region with the thinned-bone above the spinal cord at T10-11 for the induction of PT. Notice the main blood vessels and its branches. Please click here to view a larger version of this figure.
Figure 2. Spinal cord blood flow measurement. (A) The setup of the spinal cord surface blood flow measurement using a laser Doppler flowmeter and stereotaxic device to position the probe. (B) Spinal cord blood flow before and after PT was measured. In this experiment, PT was induced by illuminating with a light source for 2 min with a 12% power output. The diameter of the irradiated surface was 0.75 mm and was in the middle of spinal cord. Blood flow was recorded for up to 5 min to obtain stabilized signal before PT and up to 10 min following PT. Data from each mouse was normalized to the value prior to light illumination. The graph shows the averaged value of data from 3 mice. The arrow indicates the start of PT. Please click here to view a larger version of this figure.
Figure 3. PT-induced spinal cord ischemia. (A) Photograph of a mouse placed on the microscope for induction PT in spinal cord. The position of the mouse can be adjusted in three dimensionally using the X-Y gliding stage and a Lab-Jack. Light from the 10X objective was focused on the surface of spinal cord. (B-C) Fluorescent images of the blood vessels in the spinal cord before (B) and after (C) illumination following the injection of rose Bengal. Notice the blood clot after 2 min of irradiation (C) (see arrows). Please click here to view a larger version of this figure.
Figure 4. Nissl staining of spinal cord. Nissl staining images of a series of rostral-to-caudal spinal cord transverse sections that include normal (sections 1 and 6) and PT-induced epicenter (sections 2 – 5). The mice were sacrificed 3 days after PT. Each spinal cord section is 30 μm thick. The interval between two sections is 750 μm. The dashed line in 3rd image outlines the infarct region. Please click here to view a larger version of this figure.
Figure 5. Immunostaining of NeuN, GFAP and Iba1. Fluorescent images of NeuN (A), GFAP (B) and Iba1(C) staining from normal (upper panels) and PT-injured (lower panels) spinal cord sections. The injured mouse was sacrificed 3 days after PT. The dashed lines separate the infarction regions from normal tissues. The boxed regions show high resolution images of GFAP and Iba1 expression with a scale bar of 50 µm. Please click here to view a larger version of this figure.
Movie. PT in spinal cord induced behavioral deficits. The movie shows the movement of a normal and a PT-injured mouse in a cage. Notice the dragging of both hind-limbs of the mouse with inured spinal cord, indicating paralysis of hind-limbs (paraplegia). The movie was taken 24 h after PT in the injured mouse.
In this study, we described a photothrombotic model of SC ischemia. Due to advances in genetic engineering there has been a surge in commercially available transgenic mice which has made it possible to study the impact of specific genes involved in the ischemic pathophysiology in the SC. The aim of the study was to develop a reproducible mouse model of spinal cord ischemia. Here we adapted a cortical PT model to induce SCI in mice. Following surgery the posterior spinal vein and capillaries on the dorsal aspect of mice at the level of T11 thoracic vertebra were exposed. Then RB, a commercially available photoactive dye, was injected through retro-orbital sinus route to attain desired vascular distribution. The exposed blood vessel was then irradiated by a green light to induce the formation of thrombus and later on an infarct. Our results from histological and immunostaining methods showed that PT induced an infarct in the spinal cord and reactive gliosis in the peri-infarct region. Neurological deficits like hind limb paralysis were also observed. These data suggest that PT is a suitable model for studying pathophysiology and mechanisms of cell death after SCI. The critical step in the protocol is the use of a high speed drill to thin the surface of vertebra for visualization of the blood vessel on the dorsal surface of the SC. This step should be carefully performed as application of excess pressure might cause the drill to enter spinal cavity and damage the SC. On the other hand, uneven thinning might result in improper illumination and may produce irregular infarcts. To address this issue, frequent inspection of the surface of the bone under microscope after each short step of drilling is recommended to evaluate the thickness of the bone and to further assess the use of the drill. The use of sterile saline is recommended to washout the debris as well as for the better visualization of the exposed surface. Maintaining constant asepsis during entire surgical procedure, proper post-surgical care of the animal can improve animal survivability and increase success rate of the experiments.
Our current model of PT does not require the purchase of any expensive instruments, as any laboratory that is equipped with an epi-fluorescence microscope with a light source (such as mercury lamp, metal halide lamp, or laser of 488 nm wave length) can perform this procedure. In addition, this technique provides control over the size of the infarct by adjusting the aperture size as compared to other SC ischemia models like combined occlusion of aortic, left subclavian and internal mammary artery25 and modified aortic cross clamping method26 which are complicated and are extremely invasive. In our model a high speed drill to thin the dorsal surface of vertebra for visualization was chosen as an alternative to laminectomy, method of choice by many laboratories to induce SCI. Laminectomy involves cutting of the vertebrae which may cause excessive hemorrhage due to transection of the vertebral blood vessels and this might obscure the field for imaging. Even though some protocols advise the use of cotton swabs to clear out the excessive bleeding during laminectomy it may result in compression which may cause additional injury to the SC. Further the exposed surface of the spinal cord can come in direct contact with blood and its constituents as well as sharp edges of the cut bones which can add unnecessary variability to the experiment. Using the current PT model, infarction with different size and depth can be generated by simple manipulation of the intensity of the light source, duration of exposure and area of the exposed surface. Although the current study generated ischemia on the central region of T11 in the SC, this method can also generate infarcts at different locations along rostral-to-caudal as well as lateral direction of the spinal cord, which might benefit understanding the region-specific effect of ischemia on paraplegia. On the other hand, although the illumination is on the surface of spinal cord, the light could penetrate to certain depth in the tissue and the injury can also be induced in the grey matter. As Rose Bengal is distributed in entire circulation system, if the animal species are the same, and the age and weight are similar, we expect consistent lesion will be generated as in cortical ischemia induced by PT.
The other major advantage of PT-induced ischemia is very low mortality of animals. Low mortality means long term survival studies can be carried which might be useful in unraveling the temporal effect of ischemic injury on survival and motor function recovery. This model can also aid in understanding the cellular repair mechanisms which usually occur late in the chronic phase of injury14,19,27-29. This model also produces considerable motor function deficits which can be used to assess the efficacy of neuroprotective agents on functional recovery. In addition, this model will also allow the study of pathological changes after SCI such as axonal degeneration and regeneration, neuronal and astrocytic Ca2+ signaling and overloading in live mice using two-photon microscopy.
Like all other SCI models, PT is not devoid of drawbacks. The disadvantages of this technique are similar to those seen in cortical PT. Few of the shortcomings include lack of an anatomically clear ischemic penumbra, which is the target of many neuroprotective drugs, and the absence of reperfusion. It is well known that reperfusion following ischemia is characterized by changes like increased production of reactive oxygen species, infiltration of inflammatory cells and increased cytokine production30-32. Lack of reperfusion in PT means the changes associated with reperfusion injury in SC will remain difficult to study using this model. However, the advantages of using PT induced ischemia outweigh the disadvantages and this technique provides the researchers with an easy to perform and highly reproducible model of generating SCI in mice.
The authors have nothing to disclose.
This work was supported by the National Institutes of Health [Grant no. R01NS069726] and the American Heart Association Grant in Aid Grant [Grant no. 13GRNT17020004] to SD.
Rose Bengal | Sigma-Aldrich | 330000 | 20 mg/ml in sterile saline |
C57Bl6/J | Jackson lab | 664 | 22-25g |
Ketamine | VEDCO | NDC-50989-996-06 | 100 mg/ml |
Xylazine | VEDCO | NDC-50989-234-11 | 100 mg/ml |
Betadine solution | Purdue | NDC-67618-150-01 | 10% povidone iodine topical solution |
Normal saline | Abott Laboratories | 04930-04-10 | For diluting RB, anaesthesia and for preventing tissue from drying |
Artificial tears ointment | Rugby | NDC-0536-6550-91 | 83% white petrolatum |
Ethanol | Decon labs.Inc | 2716 | 70% ethanol for disinfection |
Metal halide lamp | EXFO, Canada | X-Cite 120 PC | Set power at 12% |
Spring scissors | Fine Science Tool | 15000-10 | for minor dissection |
Scissors (angled to side) | Fine Science Tool | 14063-011 | No. 3 handle |
Standard scalpel | Fine Science Tool | 10003-12 | for removing muscle |
Scalpel blade | Feather | 2976 | No. 10 |
Forceps (curved) | Fine Science Tool | 11150-10 | for holding tissue |
Forceps (straight) | Fine Science Tool | 11151-10 | for holding tissue |
Needle holder | Fine Science Tool | 12002-12 | for suturing |
Tissue adhesive glue | 3M Vetbond | 1469SB | to adhere to edges of the cut skin |
Monofilament polypropylene | USSC Sutures | VP-521 | Size = 4-0 (for fascia) |
Perma-hand silk | Ethicon | 683G | Size = 4-0 (for skin) |
Micro drill | Roboz Surgical Instrument Co. Inc. | RS-6300 | with bone polishing drill bit |
Laser doppler flowmeter | Moor Instruments | moorVMS-LDF1 | for monitoring change in blood flow |
Heating pad | Fine Science Tool | 21052-00 | to prevent hypothermia |
Lab-Jack | Fisher scientific | 14-673-50 | 4×4 in plate to adjust the height of the animal |
X-Y gliding stage | Amscope | GT100 | for positioning the animal under microscope |