Visual Abstract
Abstract
Traumatic spinal cord injury (SCI) is a devastating neurologic condition lacking effective prognostic and treatment methods. PET imaging of synaptic vesicle glycoprotein 2A (SV2A) has been used in measuring synapse changes. We explore the feasibility of using [18F]SynVesT-1 PET too detect the synaptic changes in a rat model of SCI. Methods: [18F]SynVesT-1 PET scans were performed on rats with T7 moderate contusion injury (n = 9) and sham controls (n = 7) on day 1 and days 9-11 after injury. The simplified reference region method 2 was used to compute the distribution volume ratios (DVRs) for the spinal cord (SC) and the brain, with the cervical cord and brain stem as the reference region, respectively. The averaged SUV ratio 30-60 min after injection was calculated as an alternative outcome measure. Diffusion tensor imaging (DTI) was used to evaluate axonal changes on post mortem SCs. Western blotting,immunohistochemical staining,and immunofluorescence staining were used to confirm the imaging results. Results: [18F]SynVesT-1 showed the highest uptake in the cervical SC. notably, the DVR at the injury epicenter in SCI rats showed a 61% decrease on day 1 and a 53% decrease on days 9-11, compared with sham controls. The changes in SUV ratio 30-60 min after injection were consistent with the changes in DVR. The fiber damage in the epicenter was identified by DTI, and the loss of SV2A was confirmed by immunohistochemical staining and Western blotting. Further, the amygdala, limbic insular cortex, and cerebellum were found to be substantially affected by the SCI on day 1 by PET. The DTI analysis revealed fiber damage in the internal capsule and somatosensory cortex. Conclusion: [18F]SynVesT-1 PET effectively identified synapse loss in the contusion SCI rat model. The quantification of synaptic density through SV2A PET presents a promising objective metric for evaluating novel therapeutics for SCI.
The estimate of annual incidence of traumatic spinal cord injury (SCI) is 54 cases per 1 million people, and approximately 308,600 people in the United States live with SCI (National Spinal Cord Injury Statistical center). Clinical outcomes vary based on lesion severity and location, possibly leading to partial or complete loss of sensory or motor function below the injury level.In this study, we used the newly developed 18F-labeled SV2A radiotracer, [18F]SynVesT-1 (20, 26-28), to assess changes in synaptic density in a rat model of T7 contusion (17). PET imaging findings were compared with ex vivo diffusion tensor imaging (DTI) and molecular biologic analyses.
MATERIALS AND METHODS
Radiochemistry
The radiochemistry procedure to produce [18F]SynVesT-1 has been described previously (26) and is briefly outlined in the supplemental materials (available at http://jnm.snmjournals.org).
Animals
Sprague-Dawley rats (female, 12 wk old, n = 16) purchased from Charles River were housed under a 12-h light/dark cycle with ad libitum food and water at the Yale animal Resource Center. All experiments were conducted with approval from YaleS Institutional Animal Care and Use Committee. Of the 2 cohorts (16 in total; Fig. 1; Supplemental Table 1), 9 rats underwent SCI procedures at the T7 vertebra using a Multicenter Animal Spinal Cord Injury Study impactor (10 g, 25 mm), as previously described (17), and 7 rats underwent laminectomy surgery as sham controls. Rat bladders were expressed 2 to 3 times daily throughout the study, and cage-side observations were made to monitor overall health and disease progression.(FIGURE 1.) (A) Diagram of experimental design. (B) Left: representative [18F]SynVesT-1 PET summed SUV30-60 min images for bas### DTI and Analyses
The brains and scs of the rats (SCI,*n* = 4; sham control,*n* = 3) were extracted for *ex vivo* DTI and analyses. The detailed procedures and analyses for multiple parameters (34), including fractional anisotropy (FA), apparent diffusion coefficient (ADC), mean diffusivity (MD), λ-parallel (λǁ), and λ-perpendicular (λ⊥), are described in the supplemental materials (35).
### Statistical Analyses
All data are presented as mean ± SD. An unpaired Student *t* test was used to compare the difference between the SCI and sham control groups in the experimental outcomes of SUV, SUVR, DVR, ADC, FA, λ‖, λ⊥, MD, and fiber tractography. Where indicated, 1-way ANOVA was used to compare the difference in the protein expression level among diffrent SC segments and in SUV30-60 min among the sham group and the SCI group at day 1 and days 9-11 after injury. A *P* value of less than 0.05 without correction for multiple comparisons was considered statistically significant.### [18F]SynVesT-1 PET Imaging Detects SV2A Loss in SC
[18F]SynVesT-1 with greater than 99% radiochemical purity was synthesized according to published procedures (27). For each animal,the injected cold mass of SynVesT-1 was 0.08 ± 0.09 µg/kg. [18F]SynVesT-1 exhibited SV2A-specific uptake in the SC of both sham control and SCI rats (Fig. 1B), which was blocked by 78% by levetiracetam (Supplemental Figs.1A and 1B).The heterogeneous tracer uptake along the axis of the SC, with cervical and lumbar regions showing higher signal than the thoracic region (fig. 1B), indicated uneven expression of SV2A in the SCs. Representative SUV“`html
Regional DVR and SUVR Changes in the SC After SCI
Regional analysis of DVR revealed a significant decrease in SC3 in the SCI group compared with the sham group on both day 1 (sham: 0.84 ± 0.03; SCI: 0.44 ± 0.08; P < 0.05) and days 9-11 (sham: 0.83 ± 0.03; SCI: 0.42 ± 0.06; P < 0.05) when using SC1 as the reference region (Fig. 2A). In contrast, the regions rostral (SC1 and SC2) and caudal (SC5) to the epicenter did not exhibit significant differences, except for SC4 in which we observed a small (9%) but significant lower DVR (sham: 0.77 ± 0.043; SCI: 0.70 ± 0.022; P < 0.05) on days 9-11 (Fig. 2A) when using SC1 as the reference region.A trend of lower DVR in SC4 on days 9-11 was also seen when using either BS or WB (Supplemental Figs. 3B and 3D) as the reference region.
We next proceeded to calculate the summed SUVR30-60 min imaging window for the SC using SC1 (fig. 2B; Supplemental Table 3), BS, and the WB (Supplemental Fig. 4) as reference regions. We also performed correlation analysis of SUVR30-60 min and DVR to explore if SUVR averaged 30-60 min after injection was a suitable surrogate for DVR (Supplemental fig. 5). We found that SUVR30-60 min is well correlated with DVR nonetheless of the choice of reference region (SC1, R2 = 0.96; brain stem, R2 = 0.96; whole brain, R2 = 0.95).
On both day 1 and days 9-11, SC3 exhibited a 48% (sham, 0.69 ± 0.02; SCI, 0.36 ± 0.07; P < 0.05) and 50% (sham, 0.72 ± 0.04; SCI, 0.35 ± 0.08; P < 0.05) lower SUVR30-60 min when using SC1 as the reference region (Fig. 2B) and using BS as the reference region (day 1: 48% difference; sham, 0.31 ± 0.003; SCI, 0.16 ± 0.03; P < 0.0001; days 9-11: 48% difference; sham, 0.31 ± 0.007; SCI, 0.16 ± 0.03; P < 0.0001) (Supplemental Figs. 4A and 4B). Similarly, when using the WB as the reference, SC3 SUVR was lower in the SCI group than in the sham group on both day 1 (44% decrease; sham, 0.30 ± 0.01; SCI, 0.16 ± 0.03; P = 0.0003) and days 9-11 (45% decrease; sham, 0.29 ± 0.02; SCI, 0.16 ± 0.03; FIGURE 4. (A) Illustration of tissue sampling sites for immunohistochemical (IHC) staining and Western blotting as correspondent to tissue locations defined using PET SUV images. (B) Representative SV2A capillary Western blotting image and quantification (n = 3, t test). (C) Top panel: IHC of neurofilament protein (NF) in the normal rostral site (left) adjacent to the epicenter (right). Bottom panel: TUNEL staining shows increased cell death in epicenter. (D) SV2A immunofluorescence staining in transaxial orientated SCs as correspondent to longitudinal IHC. Blue arrows indicate decreased signal in dorsal laminae. Red arrows indicate autofluorescence due to cell death.
To assess the impact of SCI on the brain, we extended our analysis to examine the SUV time-activity curves, dvrs, SUVR30-60 min, and the summed SUV30-60 min in selected brain regions (figs. 5A and 5B; Supplemental Fig. 9). There was no difference in the summed SUV30-60 min between the sham control and SCI group in all regions analyzed. However,we found significantly lower DVRs in the amygdala (11%,P < 0.05) and whole cerebellum (9%, P < 0.05) on day 1 for the SCI group when the BS was used as the reference region (Fig. 5B); though, the DVRs in the amygdala (18%, P < 0.01),limbic insular cortex (12%,P < 0.05), and WB (9%, P < 0.05) were lower on day 1 for the SCI group, compared with the sham group, using SC1 as the reference region. Using WB as the reference region, only DVRs of the amygdala (10%, P < 0.05) and motor sensory cortex (2%, P < 0.05) were found to be lower on day 1, whereas the DVR of BS on day 1 was slightly but significantly higher (2%, P < 0.05) on day 1, compared with the control group (Supplemental Figs. 10A-10D). We next proceeded to calculate SUVR
In our study, we used an established rat T7 contusion model, as consistent spontaneous behavioral and functional improvements were observed over time, with Basso, Beattie, and Bresnahan scores significantly increased from day 10 onward compared with day 1 in previous studies (17). Interestingly, we found the tracer uptake increased slightly on days 9-11 (DVR, −53%) compared with day 1 (DVR, −61%) in the epicenter. additionally, an increase in Ki-67 protein expression, which is a cell proliferation biomarker, was observed via immunohistochemistry (Supplemental Fig. 13), suggesting possible neurogenesis at the epicenter.To refine the contusion injury region for PET analysis,we segmented the rat SC into 5 longitudinal ROIs. The tracer uptake was found to be heterogeneous along the SC, aligning with findings in humans and monkeys (36). Notably, SC1 SUV30-60 min were similar in sham and SCI groups, indicating SC1 as a potential reference region for calculating DVR and SUVR for SC analysis. Interestingly, a similar study using [11C]UCB-J also used C3 as the reference region for analyzing tracer uptake in their C5 injury model (24). In the study, [11C]UCB-J PET was conducted to detect decreased SV2A in the rat SC.They reported reduced signals at both the epicenter (C3) and caudal levels (C6 and C7), which is consistent with our findings of decreased uptake in T8-T12, caudal to the T7 injury site. Though, [11C]UCB-J has limited clinical potential as of low specific signals in human SC and a short half-life. In contrast, [18F]SynVesT-1 has proven useful in various clinical settings and offers better prospects for clinical translation. In addition, we exploratively used BS and WB as reference regions to confirm the decreased DVR and SUVR30-60 min for the epicenter but with no changes in the caudal region (SC4). This highlights the importance of selecting appropriate reference regions for SCI PET imaging analysis.
SCI disrupts interaction between the brain and the body, leading to diverse motor, sensory, cognitive, emotional, and autonomic deficits. Significant brain changes occur during the acute to subacute phases of SCI, characterized by inflammation, altered connectivity, neuroplasticity, and potential atrophy. Our PET analysis revealed that the amygdala was the moast impacted brain region during these phases, aligning with its function in emotional regulation after SCI. Previous studies showed that SCI directly influences the amygdala through the spino-ponto-amygdaloid pathway, leading to neuroplasticity (37, 38). This change may account for the lower SV2A PET signal observed in the amygdala of SCI rats,but further molecular biologic studies are needed to explore the underlying mechanism. The cerebellum exhibited acute phase changes but seemed to adapt over time,showing no alterations in the subacute phase. collectively, [18F]SynVesT-1 PET scans effectively detected synaptic changes in the brains of SCI rats.
Although PET scans indicated synaptic loss in the amygdala from acute to subacute phases, DTI did not identify corresponding fiber changes. This discrepancy may arise becuase the changes detectable by DTI might occur later than SV2A changes identifiable by PET. Alterations in SV2A levels in the amygdala may manifest early due to stress or shifts in neural connectivity, warranting further investigation. Additionally, DTI may overlook subtle structur
Assessing Synaptic Integrity After Spinal Cord Injury: A Novel Imaging Approach
Spinal cord injury (SCI) often leads to significant neurological deficits due to the disruption of neural pathways and, critically, the loss of synapses – the junctions between nerve cells where communication occurs. Accurate assessment of synapse loss is crucial for understanding disease progression and evaluating potential therapeutic interventions. Recent research explores the potential of [18F]SynVesT-1 positron emission tomography (PET) imaging as a sensitive tool for detecting this synaptic damage, both at the injury site and in distant brain regions.
Detecting Synapse loss with [18F]SynVesT-1 PET Imaging
A recent study investigated the ability of [18F]SynVesT-1 PET imaging to quantify synapse loss in an animal model of SCI. The researchers focused on determining if the imaging agent could detect reductions in synaptic density at the epicenter of the spinal cord injury,and also in specific brain areas known to be affected by SCI. Standardized Uptake Value Ratio (SUVR) measurements, which indicate the concentration of the imaging agent, were used to assess synaptic density.
The findings revealed a significant decrease in SUVR at the injury site.Specifically,the study demonstrated a 58% and 52% reduction in [18F]SynVesT-1 uptake on day 1 and days 9-11 post-injury,respectively,when compared to control animals [[3]]. This suggests that [18F]SynVesT-1 can effectively identify acute synapse loss following SCI. Moreover, reduced uptake was also observed in the amygdala and cerebellum, brain regions implicated in emotional regulation and motor control, respectively, indicating that synaptic damage extends beyond the immediate injury location. Complementary diffusion tensor imaging (DTI) analysis confirmed these findings, revealing structural damage to white matter tracts within the internal capsule and somatosensory cortex. These areas are vital for sensory processing and motor function, aligning with the observed functional deficits in SCI.
Implications for Clinical Translation and Future Research
These results provide compelling evidence supporting the potential of [18F]SynVesT-1 PET imaging for clinical applications in SCI. Currently, assessing synaptic loss in humans relies on indirect measures or post-mortem analysis. A non-invasive imaging technique capable of quantifying synaptic density in vivo would represent a significant advancement in the field.
The ability to track synaptic changes over time could be invaluable for:
Early Diagnosis: Identifying synaptic damage in the acute phase of injury could help predict long-term functional outcomes.
Prognosis: Monitoring synaptic recovery or continued loss could provide insights into the effectiveness of therapeutic interventions.
* Treatment Growth: [18F]SynVesT-1 PET imaging could serve as a valuable tool for evaluating the neuroprotective or regenerative potential of novel therapies aimed at preserving or restoring synaptic connections.
While further research is needed to validate these findings in human SCI patients,the current data strongly suggests that [18F]SynVesT-1 PET imaging holds promise as a translational biomarker for improving the diagnosis,prognosis,and treatment of spinal cord injuries. The development of more sensitive and specific imaging agents remains an active area of investigation, with the goal of providing clinicians with the tools needed to optimize patient care and improve long-term outcomes.