|Home | About | Journals | Submit | Contact Us | Français|
Thorough preclinical testing of central nervous system (CNS) therapeutics includes a consideration of routes of administration and agent biodistribution in assessing therapeutic efficacy. Between the two major classifications of administration, local vs. systemic, systemic delivery approaches are often preferred due to ease of administration. However, systemic delivery may result in suboptimal drug concentration being achieved in the CNS, and lead to erroneous conclusions regarding agent efficacy. Local drug delivery methods are more invasive, but may be necessary to achieve therapeutic CNS drug levels. Here, we demonstrate proper technique for three routes of systemic drug delivery: intravenous injection, intraperitoneal injection, and oral gavage. In addition, we show a method for local delivery to the brain: convection-enhanced delivery (CED). The use of fluorescently-labeled compounds is included for in vivo imaging and verification of proper drug administration. The methods are presented using murine models, but can easily be adapted for use in rats.
Although systemic delivery of drugs is not the most efficient approach for achieving high drug concentrations in the CNS, systemic deliveries are convenient and well accepted by patients. Here, we will demonstrate proper procedures for three routinely used systemic administration approaches: intravenous injection, intraperitoneal injection, and oral gavage.
A reflux-resistant CED cannula for rodents is not yet commercially available. Here we will demonstrate a method for cannula construction that was adapted from a method first described by Krauze et al (Krauze 2005).
Fluorescently-labeled infusate can be imaged following CED administration, and can be monitored for changes in signal intensity as well as signal location. Typically, it is best to wait 2-3 hours after the infusion to image, so as to allow the mouse to recover from the infusion.
A lack of adverse reaction to administration of therapy is an important indicator of successful injection. For example, following tail vein injection there should be no change in appearance (e.g., size, color) of the tail. A bubble or blister following tail vein injection would indicate subcutaneous, rather than intravenous, delivery of therapy. For intraperitoneal injections, a bump on the skin or discoloration of the abdomen may indicate subcutaneous injection or damage to internal structures. In oral gavage, a mouse with labored breathing or coughing may indicate that fluid was injected into the lungs, rather than into the stomach.
For CED, neurologic function is important for assessing successful administration of therapy. A mouse that is exhibiting seizures or hemiparesis may have received an improper infusion. If a fluorescent infusate is used, in vivo imaging can be applied for assessing successful administration (Figure 3). If the infusate is not localized to the site of injection, the infusion was not successful.
Figure 1: CED Cannula and Surgical Set-up. The basic elements of the convection-enhanced delivery surgical set-up are shown. A microinfusion pump (A) is attached to the infusion cannula (B, shown enlarged at top). A stereotaxic frame (C) is used to position the probe. This image does not include heating and anesthetic equipment also used during the procedure.
Figure 2: Mouse Skull Suture Lines. The CED infusion site (red star) can be located by identifying the intersection of the sagittal and coronal sutures (the bregma) and then measuring 2mm lateral and 1mm posterior of the bregma.
Figure 3: Representative Results from Successful CED. Liposomes labeled with a far-red fluorescent dye were infused into the mouse brain by CED and imaged both in vivo and ex vivo. A successful infusion shows a fluorescent signal localized to the site of infusion both in vivo (A) and ex vivo (B). The signal should be localized to the infused hemisphere without leakage into the contralateral hemisphere.
Figure 4: Representative Image of Successful CED into the Rat Brainstem. Fluorescently-labeled liposomes were infused into the rat brainstem. In rats, as much as 20μL can be infused. The increased thickness of the rat skull and depth of injection precluded in vivo imaging, but correct infusion location could be verified by ex vivo imaging of the dissected brain.
Preclinical evaluation of the efficacy of any therapeutic agent must take into account pharmacological properties of the agent and the target tissue of interest. While systemic methods of administration are in general more facile and better tolerated by patients, the selectivity of the blood-brain barrier often necessitates local delivery of therapy for treating CNS disease. Here, we have demonstrated one method of direct delivery to the brain: CED. Other forms of local delivery to the brain include direct administration to the cerebrospinal fluid, intratumoral bolus injection, and ventricular injection. For these methods, diffusibility of therapeutic is critically important, with poor diffusion limiting the region of coverage to a few millimeters from the injection site (Jain 1989, Bobo 1994). In contrast, CED uses positive pressure to increase the area of drug distribution (Bobo 1994), while still enabling therapeutic targeting to specific neuroanatomical structures. In our laboratory, we have successfully performed CED into the brainstem as well as into the caudate putamen in rodents (Figure 4).
The ability to conduct CED in rodent models has become increasingly important in association with increasing interest in maximizing therapeutic efficacy when treating various types of CNS disease. CED can be used to deliver a variety of agents, including purified proteins, small molecule drugs, and viruses (Gill 2003, Degen 2003, Szerlip 2007). The range of diseases that can be treated with CED includes cancers of the CNS (Yamashita 2007) as well as neurodegenerative diseases such as Parkinson's disease(Gill 2003). As CED research continues to expand, the need for in vivo imaging of CED administered therapy is similarly increasing. While the fluorescent imaging demonstrated here would likely not be visible through the human skull, other methods employing co-infusion of MRI contrast agents (Dickinson 2008) are attracting interest for assessing applicability for monitoring infusates in patients with CNS disease.
All procedures demonstrated here were approved by the UCSF Institutional Animal Care and Use Committee.
NS65819 (CDJ, TO), NS049720 (CDJ), CA097257 (CDJ, TO), CIRM DR1-01426We would like to thank Raquel Santos for technical assistance.