The cisterna magna (CM) is the largest subarachnoid cistern between the arachnoid and pia mater layers; it receives cerebrospinal fluid (CSF) from the fourth ventricle [1]. The CSF continuously circulates in the cerebral ventricles, subarachnoid space, and spinal cord central canal while providing nourishment and protection to, and enabling waste removal from, the brain [2]; CSF collection and intracisternal injection via suboccipital puncture are convenient strategies because they are less invasive and show a minimal risk of contamination [3]. Therefore, numerous studies using intracisternal administration of biologics, including adeno-associated virus (AAV)-mediated gene delivery and injection of antisense oligonucleotides (ASOs), recombinant enzymes, and novel agents as promising therapeutic candidates for neurodegenerative diseases, have demonstrated the medical efficacy of this technique and pathological mechanisms underlying various neurological diseases [4-6].

Intracisternal injection via suboccipital puncture with a manual stereotactic (non-stereotactic) and stereotactic method is widely used for drug distribution through CSF flow into the brain [7], since this method has several advantages. Although this route has diverse advantages for direct transport into the brain, the low accuracy of intracisternal injection remains a hurdle in the development of safe and reliable techniques for drug delivery. Intracisternal injection with a non-stereotactic method in particular requires anatomical knowledge and its efficacy depends on the skills and experience of the technician because otherwise, several side effects, including blood contamination, CSF leakage, and brain stem damage, can occur [8-10].

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by memory and cognitive decline [11]. Accumulation of amyloid-beta plaques and the formation of neurofibrillary tangles in the cerebral cortex and hippocampus are regarded as typical pathological hallmarks of AD [12, 13]. Furthermore, several abnormalities in brain glucose metabolism have been observed in AD patients. Cerebral impairment of insulin signaling and translation of insulin receptors have been recognized as early signs of AD progression [14-17].

Streptozotocin (STZ) is a diabetogenic compound generally used to establish animal models of diabetes owing to its ability to selectively impair the insulin signaling pathway [18]. Intracerebroventricular (ICV) administration of STZ disrupts the homeostasis of brain insulin signaling and defects in cerebral glucose metabolism [19]. This is accompanied by behavioral, neuropathological, and biochemical changes similar to those observed in the pathology of AD [20-22]. However, this technique is invasive and greatly limits the generalizability of the results obtained in monkeys. Although this technique is invasive method for studying the brain pathology in monkeys, there are some technical challenges and limitation to overcome unanticipated adverse effects. The ICM administration of STZ is also carried out, but much less frequently than ICV approach.

Therefore, the present study aimed to determine the effects of STZ administration in cynomolgus monkeys via the intra-cisterna magna (ICM) route using a novel technique for X-ray-based real-time three-dimensional imaging coupled with the c-arm cone beam computed tomography (CBCT) technology, XperCT (Philips), which can assess three-dimensional images of soft tissue, and bone structure; to the best of our knowledge, our study is the first to report the use of the XperCT stereotactic system for injecting drugs into the brain tissue to establish a monkey model of AD. We suggest a new alternative method for the stable and reproducible delivery of drugs or molecules in the brains of non-human primates (NHPs).

Ethical statement

The experimental procedures using experimental animals were approved by the Institutional Animal Care and Use Committee of the Korea Research Institute of Bioscience and Biotechnology (KRIBB) Institutional Animal Care and Use Committee (Approval No. KRIBB-AEC-20253). The experimental procedures were performed in accordance with the national guidelines and in compliance with the Guide for the Care and Use of Laboratory Animals.

Experimental animals

Cynomolgus monkeys (Macaca fascicularis) were obtained from Suzhou Xishan Zhongke Laboratory Animal Co. (Suzhou, China); they were housed in individual indoor cages at the National Primate Research Center (NPRC) of the KRIBB, as described previously [23]. Cynomolgus monkeys were fed twice with commercially available monkey feed (Harlan, USA), supplemented with various fruits and water ad libitum. The health of the monkeys was monitored by the attending veterinarian, consistent with the recommendations of the Weatherall Report. Animal health monitoring for monkeys was performed via microbiological tests, including those for B virus, simian retrovirus (SRV), simian immunodeficiency virus (SIV), simian virus 40 (SV40), and simian T-cell lymphotropic virus (STLV), as described previously [24]. The animals were housed in a temperature-controlled room (24±2℃) with 50±5% humidity and a 12-h light/dark cycle.

Intra-cisterna magna (ICM) injection of STZ and CSF collection

The cisterna magna is located below the cerebellum, with an intact atlanto-occipital joint (Fig. 1). STZ (2 mg/kg; Sigma-Aldrich, St. Louis, MO, USA) was dissolved in artificial CSF (aCSF; Harvard Apparatus, Holliston, MA, USA). The custom-built ICM-stereotaxic frame used in the present study was designed for compatibility between CT and MRI (Fig. 2). All monkeys were initially anesthetized via the intraperitoneal injection of a cocktail mixture of ketamine (5 mg/kg) and atropine (0.02 mg/kg) and fixed in the sphinx position using a custom-built ICM-stereotaxic frame for image-guided stereotactic system under isoflurane-induced anesthesia (1.5% in 2 L/min oxygen). After confirmation of the correct needle tip position within the cisterna magna using XperCT imaging, CSF was extracted to verify the accurate position of the cisterna magna. Subsequently, the tube with the needle tip was connected to a Hamilton syringe containing STZ; ICM injections were administered once every week for 4 weeks.

MRI

MRI experiments were performed using a 3.0-T MRI scanner (Achieva 3.0T, Philips Medical Systems, Best, Netherlands) with an 8-channel knee coil. Three-dimensional (3D) sagittal T1-weighted images were acquired using the turbo field echo sequence with the following settings: TR/TE=14/6.8 ms, 128×128 field-of-view (FOV), matrix size 256×256, voxel size 0.5×0.5×0.5, and number of slices=150. The details of the MRI protocols were the same as those described in a previous report [25].

Blood and CSF analysis

Peripheral blood and CSF samples from all cynomolgus monkeys were collected, and cell type composition was analyzed using a hematology analyzer (Hemavet950; Drew Scientific, Miami Lakers, FL, USA).

Perfusion and tissue preparation

Two weeks after the final STZ injection, all cynomolgus monkeys were transcardially perfused with 500 ml of 100 mM phosphate-buffered solution (PBS) under deep anesthesia with the intramuscular injection of a cocktail of ketamine (5 mg/kg) and atropine (0.02 mg/kg). Whole brains were removed from the skull, washed with cold PBS, and separated bilaterally. Hippocampal proteins were harvested from monkey brains using punches on 4-mm-thick slices.

Western blot analysis

Protein samples of cynomolgus monkey hippocampi were harvested using the PRO-PREP protein extraction solution (Intron Biotechnology, Seongnam, Korea). Equal amounts (15 μg) of protein were separated by electrophoresis on 10~15% SDS-PAGE gels and transferred onto nitrocellulose membranes (BD Biosciences, Franklin Lakes, NJ, USA). The membranes were blocked by incubation with blocking buffer (BD Biosciences) and probed with the following antibodies overnight at 4℃: anti-β-actin, anti-GFAP (Sigma-Aldrich, St. Louis, MO, USA), anti-NeuN, anti-synaptophysin, anti-PSD95, anti-phospho(p)-tau(S262), anti-p-tau(T181), and anti-p-tau(S396) (Abcam, MA, USA). Next, the membranes were washed with TBS saline containing 0.1% Tween-20 (TBST) and incubated with horseradish peroxidase-conjugated secondary antibodies (Cell Signaling, MA, USA) for 1 h at room temperature. After washing with TBST, the specific binding was detected using a chemiluminescence detection system (Thermo Scientific, MA, USA).

Statistical analysis

The data represent the mean±SD from three independent experiments (n≥3). Experimental differences were tested for statistical significance using two-way ANOVA conducted of variance using GraphPad Prism 9 software (San Diego, CA, USA). Statistical significance was set at p<0.05, and is indicated on the graphs using asterisks; p-values<0.01 and <0.001 were indicated by two and three asterisks, respectively.

T1-weighted MRI images obtained after the intracisternal tracer injection reveals the CSF dynamics

To develop a novel method for intracisternal injection using the XperCT-guided stereotactic system, we designed a protocol that enables the assessment of real-time updated images during the intervention. The animals were fixed in the sphinx position using a custom-built ICM-stereotaxic frame under anesthesia (Fig. 3A). After registration with the preoperative XperCT reconstruction, the planning and guiding needle insertion can be acquired to establish the path to the target site, i.e., the cisterna magna (CM). To verify the real-time position of the needle after insertion, we obtained a new XperCT image and confirmed the target point on the postoperative imaging (Fig. 3B). A CSF tracer was used to visualize the intracisternal injection through the target point. The CSF tracer (100 mM) was slowly injected (25 μl/min) into the CM, and brain T1-weighted MRI was performed after injection. Representative MRI images of tracer infusion into the CM showed the uptake of CSF into the brain parenchyma (Fig. 3C). These results suggest that the CSF tracer (injected via the ICM route) flows along the CSF circulation pathway and moves along the basement membrane of the brain parenchyma.

Effects of STZ administration (via the ICM route) on AD pathology-associated features in hippocampus of cynomolgus monkeys

To apply the XperCT-guided ICM administration system, we selected STZ, which is known to trigger AD pathology in cynomolgus monkeys [25-27]. We injected aCSF (vehicle group; n=2) and STZ (2 mg/kg; STZ group; n=2) using XperCT-guided ICM at weeks 1, 2, 3, and 4 (Table 1). Two weeks after the last ICM-STZ administration, all monkeys were sacrificed and further protein analysis was performed (Table 2 and Fig. 4A). Given that several studies have suggested that STZ successfully triggers AD pathology-associated features, such as neuronal loss, synaptic loss, and abnormal tau phosphorylation [28-31], we analyzed neuronal loss, synaptic loss, and abnormal tau phosphorylation by immunoblotting analysis. STZ-induced neuronal loss was confirmed by assessing the levels of the NeuN protein, a neuronal marker. The NeuN protein levels decreased significantly in the STZ-injected group (Fig. 4B). In addition, we determined the expression levels of the pre- and post-synaptic proteins synaptophysin and PSD95, respectively. The expression levels of synaptophysin and PSD95 in the hippocampus decreased in the mice from the STZ-injected group (Fig. 4C). We also verified that the phosphorylation the of tau epitopes, i.e., the levels of p-Tau(S262), increased significantly in the mice from the STZ-injected groups (Fig. 4D), whereas the levels of p-Tau(T181) and p-Tau(S396) seemed to be more affected by individual differences (Fig. 4E). These results indicate that diverse AD pathology-associated features were observed in the hippocampus of cynomolgus monkeys injected with STZ via the ICM route.

The ICM route of administration has been continuously developed for enhanced central nervous system (CNS) drug delivery. This method is widely used to bypass the blood-brain barrier and has distinct advantages for direct delivery into the CNS [32]. An alternative to CSF-mediated delivery routes is lumbar intrathecal (IT) injection and ICV infusion. However, IT injection via lumbar puncture and ICV infusion via cranial puncture are challenging because of procedural complexity, risk of possible contamination, and postoperative complications [33-35]. Lumbar puncture used to obtain CSF or for chemotherapy is a highly skilled procedure that requires practical experience and specific knowledge of the relevant anatomy. In addition, cranial puncture site infections and intracerebral hemorrhage after ICV infusion are inherent surgical risks. Therefore, ICM administration has a clear advantage, given that it is widely used in animal models. In addition, the ICM route has been used extensively, particularly in NHPs, because it offers the easiest entry into the ventricles of the brain and subarachnoid space around the brain and spinal cord, except for craniotomy [36].

Although there are infectious and non-infectious complications related to ICM injection, they can be controlled and prevented by improving the accuracy of ICM injection using advanced radiologic techniques such as X-ray, CT, and MRI [3, 37, 38]. Here, we describe the design of an XperCT-guided stereotactic system to optimize high-precision and stable CSF collection without multiple complications. To achieve X-ray-based real-time 3D imaging during the ICM targeting procedure, we used a newly developed system, the XperCT technology. The XperCT-guided technique allows precise planning of access to ensure injection of the drug at an appropriate position to avoid brainstem injury [39]. Therefore, the XperCT-guided ICM injection technique described herein provides evidence to support the wide applicability of the stereotaxic procedure of NHPs.

NHPs, including cynomolgus monkeys (Macaca fascicularis), share brain structure and function and are genetically similar to humans [40]. Moreover, AD-related natural processes in the brains of some aged NHPs are similar to those in humans [41-43]. Therefore, there is a need for an NHP-based model that reflects the AD pathology-associated features observed in humans. To date, several groups have developed transgenic monkey models expressing human mutant transgenes characteristic of the familial form of AD, such as the amyloid-β precursor protein or Tau protein [44, 45]. Although these transgenic models are valuable tools for studying the molecular mechanisms underlying AD pathogenesis, they do not fully replicate certain aspects of the neuropathology of AD. Therefore, non-transgenic AD models have become an important focus of research. Administration of STZ via the ICV route causes the establishment of an insulin-resistant brain state, a prominent representative feature of non-transgenic AD models; this feature has been proposed as a well-characterized, and is considered an appropriate, AD pathology-associated feature [19-22, 29]. However, a major disadvantage of ICV injection is that it is an invasive procedure involving a skin incision and trauma to the brain tissue. Earlier studies have confirmed that accurate and reproducible access to the CSF using ICM injection ensures the diffusion of the drug into the parenchyma and is similar to the diffusion observed after ICV injection [46, 47]. Therefore, we previously developed a clinically relevant STZ-administered AD model using ICM injection in monkeys and rats, which is characterized by cerebral damage, disintegration of the neurovascular unit, neuroinflammation, amyloid deposition, neuronal loss, and tau phosphorylation [25, 27, 48]. However, there are some technical challenges and limitations to constructing stable and reproducible AD NHP-based models, including high costs and skilled technical infrastructure of XperCT. To the best of our knowledge, the present study is the first to apply the XperCT-guided system to achieve the stable and reproducible establishment of NHP models of AD via intracisternal STZ administration using a custom-built ICM-stereotaxic frame. Pathological and molecular changes, such as neuronal loss, synaptic impairment, and hyperphosphorylation of Tau, were observed in our NHP model of STZ-induced AD. However, other typical AD hallmarks, such as astrocyte and microglial activation, between the animals in the vehicle and STZ groups were not significantly different. To overcome the limitations of our results, such as low sample size, additional experimental conditions, such as the examination of more than two monkeys per group and/or at various time points, are required. Moreover, in the monkey model of AD described in the present study, MRI scans (showing prominent ventricular dilation and parenchymal atrophy) and evaluation of cognitive function were not performed. Therefore, we believe that diverse observations, such as long-term follow-up observation of changes in cognitive function [49], brain scans using MRI [50], and AD marker screening from CSF and blood [51] are required to obtain more accurate information regarding STZ-induced AD monkey models.

Taken together, we designed and applied a new alternative method, i.e., the XperCT-guided injection method, as a stable and reproducible delivery system for administering drugs into the brains of NHPs via the ICM route. Our findings suggest that this new approach can be applied for the delivery of diverse molecules, such as drugs, ASOs, viral vectors, and disease inducers, into brain tissues.

This study was supported by the Korea Research Institute of Bioscience and Biotechnology Research Initiative Program (KGM4562222 and KGM5282221), and by the National Research Council of Science & Technology (NST) grant by the Korea government (MSIT) (No. CPS21101-100), and by the Korea Medical Device Development Fund grant funded by the Korea government (the Ministry of Science and ICT, the Ministry of Trade, Industry and Energy, the Ministry of Health & Welfare, the Ministry of Food and Drug Safety) (Project Number: 9991006929, RS-2020-KD000264).

Differential count of Leukocytes and Erythrocytes in blood and CSF from cynomolgus monkeys

Summary of NHPs in this study