| BACKGROUNDDiabetes mellitus is a metabolic disorder that affects more than 300 million people around the world. The most common complication of diabetes is sensorimotor polyneuropathy. At least a third of patients with sensorimotor polyneuropathy develop diabetic neuropathic pain (DNP), leading to a negative impact on their quality of life. Neuropathic pain is characterized by aberrant sensations, including spontaneous pain, alterations in pain perception, hypersensitivity to painful sensations (hyperalgesia) and painful sensations to innocuous stimuli (allodynia).Painful diabetic neuropathy is traditionally considered to be attributed to various metabolic and morphological changes in the peripheral nervous system. Numerous studies have demonstrated that these changes connect with symptoms of neuropathic pain. Sensory nerve fibers with peripheral nerves, for example C-fibers, which have the distal terminals of their axons close to peripherally innervated tissues, can increase spontaneous activity and become hypersensitive as a result of disease or injury. Consistent with the idea that hyperexcitability of primary afferents can engender inappropriate autonomous activity and exaggerated responses to peripheral stimuli, the expression of multiple ion channel isoforms, such as Nav1.3, Navl.8 and Navl.9, which contribute to action potential electrogenesis, has been shown to be upregulated in dorsal root ganglion neurons in both patients and animal models with neuropathic pain.However, peripheral neuropathy might only partly explain the neuropathic pain observed in diabetes. Existing pharmacological approaches partially relieve the pain by targeting the mechanisms of peripheral nerve damage or only treat the symptoms but not the cause. None of the drugs provide complete or long-term pain relief and have significant side effects. The findings about the peripheral nervous system showed that the above does not suggest that the mechanisms of peripheral neuropathy function in isolation to produce pain. Furthermore, recent studies suggest that central nervous system dysfunction also contributes to neuropathic pain.Paulson et al. found widespread increased activity within the sensory-discriminative pain system, including the ventral posterolateral (VPL) nucleus, through an autoradiographic examination that detected regional cerebral blood flow in rats with DNP. A study by Cauda et al. used functional MRI (fMRI) to investigate the connectivity of neuronal networks in the brains of patients with painful diabetic neuropathy. They found that the spontaneous component of diabetic neuropathic pain is the result of aberrant default functional connectivity. More recent studies by Fischer applied electrophysiological methods to functionally assess the firing pattern of neurons and demonstrated that thalamic VPL neurons showed increased spontaneous activity and hyper-responsiveness that led to central generation of pain signals associated with diabetic neuropathy.Furthermore, disorders of metabolism and the microcirculatory system in diabetes can directly lead to systemic nerve injury. Especially in the brain, the damage to the central nervous system is likely to result in various pathological changes associated with pain. So, the mechanism of diabetic neuropathic pain in the central nervous system may be more complex than previously thought. The exploration of aberrant nerve function and the pain conduction pathway is the foundation of pathological studies of the central nervous system with painful diabetic neuropathy. Therefore, many researchers have applied multiple measures to examine the pain pathway in the intracerebral nerve under conditions of normal physiology and abnormal pathology. They found that multiple neurons, including the medial thalamus, anterior insular cortex (AIC), lenticula, cingulate gyrus and prefrontal cortex (PFC), participate in pain signal transmission under diverse pathological conditions.Hence, the key task we are now facing is to identify the specific zones with dysfunction from the nerve pathway in painful diabetic neuropathy. Abnormal spontaneous discharge is one of the significant forms of neurologic lesions. A previous study of the firing pattern of neurons within the thalamic VPL nucleus also confirmed increased spontaneous activity in an experimental model of pain associated with diabetes. Thus, identification of the multiple abnormal regions of electrical activity in cerebral neurons or regions under DNP and exploration of their role in the initiation, regulation and maintenance of DNP will help to complete the underlying mechanism of central nervous system dysfunction that contributes to DNP. Locating these lesioned regions can provide a deeper understanding of target spots and has the potential to lead to new therapies for painful diabetic neuropathy that treat the underlying cause of the neuropathy rather than the resulting pain.The most commonly used functional MRI methodology is the blood oxygenation level-dependent technique. However, this technology is not ideal for the present study of the diabetic chronic pain state, as it requires a rapidly changing signal that is not always present during ongoing, chronic neuropathic pain. In our study, we utilized in vivo manganese enhanced magnetic resonance imaging to noninvasively determine the abnormal functional activation of cerebral regions in an experimental rat model with DNP. Manganese ion (Mn2+) is a paramagnetic MRI contrast agent that instills positive contrast in spin-lattice (T1) weighted MRI images. The continuous accumulation of Mn2+in brain regions due to neuronal electrical activity as known to enter cells via ligand-or voltage-gated calcium (Ca2+) channels during nerve action potentials. Because Mn2+efflux from cells is slow, its accumulation in abnormal spontaneous activity-induced brain regions can be measured as a specific increase in tissue T1 at high spatial resolution hours later after systemic administration in waking animals. In the present study, abnormal spontaneous activation is uncontrollable and difficult to predict; thus, the accumulation of Mn2+with time can help visualize more subtle and instantaneous dysfunction in nerve cells.OBJECT:In this study, we investigated whether dysfunctional activity within some special cerebral regions during painful neuropathy diabetes occurs by manganese enhanced magnetic resonance imaging (MEMRI).METHODS:1. Induction of diabetesThe method of inducing of diabetes in rats was outlined in a previous study and was followed with slight modifications. Briefly, diabetes mellitus was induced by a pancreatic β-cell toxin that produces diabetes. A single intraperitoneal (I.P.) injection of streptozotocin (STZ) (60 mg/kg; Sigma, St. Louis, MO) dissolved in citrate buffer (pH4.5) was administered after animals were fasted overnight to maximize the effectiveness of the STZ treatment. Age-matched control rats (n=12) were injected with equal volumes of acidic buffer vehicle only. piabetes was confirmed by fasting blood glucose level determination on samples taken from tail vein bleeds 24 h,72 h, 7 days and weekly after STZ injection using an ACCU-CHEK active blood-glucose monitoring system (Roche Diagnostics, Germany). Diabetes mellitus was defined as hyperglycemia with a fasting blood glucose level higher than 16.7 mmol/L. Before STZ injection, the rats with an abnormal fasting blood glucose level were removed. Animal body weight, urine glucose, etc., were monitored simultaneously during the experimental period.2. Diabetic neuropathic pain measurementA blinded experimenter performed behavioral assessment for tactile allodynia in a dedicated quiet room under invariant conditions on both diabetic and non-diabetic control rats. The up and down method of Dixon was used to determine the paw withdrawal mechanical threshold with mechanical stimuli applied to the plantar surface of the hind paw with Von Frey filaments, and the resultant scores were used to calculate the threshold. Briefly, prior to measurement, the animals were acclimatized in a see-through inverted plastic cage (18×15×18 cm) on a wire mesh (0.8×0.8 cm) floor to allow access to the paws for at least 10 mins. Then, the use (5-10s each) of different Von Frey filaments with increasing stiffness (0.004-15 g) with incremental forces was used to stimulate the midplantar surface of the hind paws with paw withdrawal constituting a positive response. The testing paradigm began with 2.0 g. A negative response prompted us to apply the next heavy filament; the inverse method was used for a positive response. The measurement was continued until four measurements had been taken after the initial change in behavior. Each rat was tested three times and the mean withdrawal threshold was determined. Baseline behavioral testing was performed before STZ or vehicle injections, and also weekly thereafter. Tactile allodynia was defined as a withdrawal threshold of less .than or equal to 2.0 g. According to the tactile allodynia, the diabetic rats were divided into two groups:the diabetic neuropathic pain group and the diabetic neuropathic painless (DNPL) group.3. Manganese ion (Mn2+) injectionBefore MEMRI, the contrast agent Mn2+needed to be injected into the body of the animals. The rats were anesthetized with 5% isoflurane. To maintain the appropriate body temperature of 37℃ during the infusion of Mn2+, the animals were placed on a resistive heating pad and maintained under anesthesia with 2% isoflurane in 100% O2. Manganese chloride (MnCl2; Sigma, MO), dissolved to 100 mM in isotonic saline (100 mg/kg), was constantly administered intravenously via the tail vein at a rate of 2.08 μmol/min (1.25 ml/h) using a syringe pump. After the infusion of the Mn2+, the animals usually woke approximately 1 min after ceasing exposure to isoflurane and could drink and eat independently after anesthesia recovery. Then, the rats were housed individually in darkness for 24 hours in an absolutely quiet environment and were removed from stimulation by light and sound before the MRI exam. The injections of Mn2+were also accomplished under dim red lights or darkness.4. MEMRI procedureTwenty-four hours after the infusion of Mn2+, the diabetic animals were anesthetized through a nose cone with isoflurane as described above and placed on a stereotaxic animal holder with a recirculating water blanket. The head holder was adapted with a movable bite and ear bars and positioned fixed on a magnet chair. Whole-brain MEMRI was performed on a 7.0T MRI animal scanner (Bruker, BioSpec 70/20USR; Ettlingen, Germany) equipped with a gradient insert with a 20-cm inner diameter. A 7.1-cm inner diameter whole body transmit-only coil and a matched receive-only surface coil were placed dorsal to the head. Images were acquired using a FLASH-3D T1-weighted sequence with the following parameters: repetition time (TR)= 35 ms; echo time (TE)= 4 ms; field of view (FOV)= 30x30x22 mm; flip-angle=30°; matrix= 256x256x55; number of averages (NA)= 6; no interslice gap. Body temperature, heart rate and breathing rate were monitored and kept constant throughout the procedure. In any case, unintended stimuli coming from the surrounding environment should be eliminated carefully. The whole process was performed under dim reds or darkness and ear plugs were used to block noise.5. MEMRI data preprocessingScanned Bruker 2dseq file were first transformed to ANALYZE format by using MRIUtil and saved as 16-bit unsigned integer data, which included image and head file of ANALYZE format. MRIUtil is an application for viewing MRI image data and performing post-processing operations on raw MRI image data as well as reconstructed MRI image data. It supports reading Analyze, NIFTI, Bruker 2dseq, Philips PAR/REC, GE, Siemens, and some basic raw image formats.Then,use the voxel size augmentor of dpabi toolbox to magnify the image voxel size 10 times so that suitable for subsequent registered to the WistarRatMRI template image. The original images voxel size is 0.12×0.12×0.4 mm and the template image voxel size is 1.25×1.25×1.25 mm. After the enlargement of images voxel size so that they can register each other commendably.Thirdly, the registration and smoothness were processed by SPM12. With a publicly available digital WistarRatMRI template used as the target and registered all the augmented images to this rat template. The affine transformation comprises 12 parameters (translations, shears, rotation and scale factors in three dimensions). Then, registered images are smoothed using a 4-mm full-width-half-maximum (FWHM) isotropic Gaussian kernel to eliminate the noise and improve signal-to-noise ratio.Eventually, a two-tailed Student’s t-test was performed voxel-wise on the data. We used a threshold of p=0.05 for statistical significance. With the results projected to the RatMRI template image.6. Volume of interest analysisAccording to the voxel-base statistical maps of the T1WI signal intensities in the brain comparing DNP rats to DNPL rats, a volume of interest (VOI) analysis was performed to quantify the activity variation on these special structures. Spherical VOIs with a radius of 0.5 mm were placed at the midpoint of the target structure on the bilateral brain. The anterior-posterior and medial-lateral coordinates depend on the bregma and midline and were defined from the stereotactic coordinates from the rat brain atlas. The intensity in the temporal muscle was measured with the same size of VOI to serve as a reference. The signal intensities with these VOIs in each individual sample were normalized with respect to their own reference. In both groups, a normalized intensity value within each VOI of these regions with changing activity was averaged separately.7. Statistical analysisResults are presented as mean values and standard error of the mean (S.E.M.). Statistical tests were performed at an a-level significance of 0.05 for all analyses using parametric tests. Significant within-group changes over time were performed by repeated measures ANOVA; between groups were determined by mixed-model repeated measures ANOVA. One-way ANOVA pairwise comparison and Student’s t-test were used where appropriate. All statistical analyses were performed using the software package, SPSS and other software programs listed above.RESULTS:1.Significantly enhanced activity of brain regions including the primary and secondary somatosensory cortex (SI, SII), anterior cingulate cortex (ACC), ventromedial prefrontal cortex (VMPFC), piriform cortex (Pir), inferior olive (Io), and parts of the cortexes of the amygdala and insula.2.Nevertheless suppressed intensity in regions of brain activation containing the primary motor cortex (MI) and secondary motor cortex (MⅡ).3.Normalized VOIs analysis of these regions in the DNP and DNPL groups shown consistent consequences with voxel-base statistical maps (Table.2).CONCLUSION:We detected the abnormal activation of the brains of rats with DNP compared to DNPL rats with MEMRI, an emerging functional imaging technique. We found some specific cerebral regions with varying changes in functional activity under chronic pain modulation in the rat model with DNP. Regardless of the increased or decreased activity regions detected in our research and their possible role in the initiation, regulation and maintenance of pain modulation, some mechanisms still remain unclear. Although the complex nerve transmission in DNP lead to these increased regions, the region with decreased activity, the motor cortex, must play a key role in the regulation of DNP. However, these regions may not include all the target regions due to the present sample size of our study. Furthermore, MEMRI can only be utilized in animals due to the toxicity of Mn2+at present. Although limited by the neurotoxicity of Mn+, more detailed regions may not be shown by MEMRI, necessitating the use of emerging techniques to confirm and complete this work. The location of all target regions requires further in-depth study in order to better understand the mechanism of DNP and allow new therapies for painful diabetic neuropathy. |
PDF Full Text Request