Gene vector ‘magic bullet’: targeted expression in the central nervous system after peripheral delivery using the synapsin promoter

KEYWORDS:

• Adeno-associated virus
• amyotrophic lateral sclerosis
• neuron
• promoter
• synapsin promoter

1. Introduction

Wide-scale delivery of a gene vector to the central nervous system (CNS) has expanded experimental paradigms and therapeutic interventions. A peripheral-to-central route of administration is attractive because it avoids direct invasive damage to the brain or spinal cord. The study by Foust et al. achieved efficient neuronal expression after intravenous delivery of recombinant adeno-associated virus serotype 9 (AAV9) to mice [1], which opened up novel approaches such as disease modeling for amyotrophic lateral sclerosis in rats [2,3]. AAV9 coupled with the cytomegalovirus/chicken beta-actin promoter efficiently transduces and expresses in the heart, the liver, and muscle [1,2], so a more targeted method is, therefore, urgently needed if specific expression in neurons throughout the CNS is the goal.

2. AAV-synapsin promoter targeting

A neuron-specific promoter such as the synapsin promoter should be selective for neuronal expression after a peripheral delivery. The recombinant human synapsin promoter is a popular promoter for gene transfer to the CNS. Kügler et al. incorporated a short ~500 bp sequence from the human synapsin promoter into an adeno-associated virus (AAV) vector and demonstrated a neuronal expression pattern in the brain, avoiding expression in glia [4]. The early characterizations of the neuronal specificity of the human synapsin promoter are discussed in [4]. The neuronal expression conferred by these vectors and the small promoter size have led to widespread use in the field. The synapsin promoter is often used in vectors for elegant optogenetics studies. Specific cell populations can be transduced to determine their role in specific functions. As one robust example, Alsahafi et al. recently used AAV-synapsin promoter vectors and optogenetics to control the inspiratory function of rats after focal expression of channelrhodopsin-2 in the medulla oblongata [5]. As mentioned, the synapsin promoter confers expression to neurons, and its small size yields more space for transgene design and a recombinant DNA that is within the packaging limits of the system, 4.5–5 kb.

Several studies have administered recombinant AAV-synapsin promoter vectors into the cerebral ventricles for more widespread CNS expression compared to focal injections [6,7]. Of note is the successful preclinical gene therapy in a mouse model of Fragile X Syndrome by restoring the functional Fmr1 protein in the CNS [7]. To our knowledge, only one study has applied AAV-synapsin promoter vectors to mice by an intravenous route of administration [8]. By this route of administration, Huda et al. noted neuronal expression throughout the CNS as well as expression in the heart and liver [8], so it appears that the recombinant human synapsin promoter should, therefore, be referred to as neuron-selective, not neuron-specific in the context of a recombinant AAV9 vector. More work will be needed to address some of the key remaining questions with respect to wide-scale expression derived from the peripheral delivery of AAV-synapsin promoter vectors: (1) will the neuronal expression be strong enough to be physiologically significant; (2) will AAV-synapsin promoter vector expression be stable over time; (3) can neuron-specific expression be achieved; and (4) will the combination with other vector elements improve expression and specificity? We are currently addressing these questions with AAV-synapsin promoter vectors in our laboratory since wide-scale, yet targeted expression is such an important goal.

3. Genetic-based targeting

The synapsin promoter has also been employed in Cre recombinase-dependent AAV vectors [9]. Investigators have used this technique to define which type of neuron underlies specific reward responses to cocaine. AAV vectors with floxed transgenes are focally injected into the nucleus accumbens of either D1 or D2 dopamine receptor-Cre driver mouse lines to map differential responses in these two types of striatal neurons [9]. It will be interesting to test whether this type of approach can be translated to the wide-scale technique, for example, to induce the expression of disease-related proteins in specific target cells throughout the nervous system. Combining wide-scale AAV delivery to specific neuronal Cre driver lines could facilitate and expedite hypothesis testing compared to crossing two or more transgenic lines together.

Xie et al. described a microRNA sequence that efficiently dampens the peripheral expression in liver for the goal of targeting the CNS [10]. It will be interesting to combine this microRNA into a synapsin promoter vector to test if fully exclusive neuronal expression can be achieved. Another primary targeting method will be modifications of the AAV capsid by selection assays that recover particular engineered variants that have enhanced efficiency to transduce neurons [11]. Again, it will be interesting to combine these new variants with other neuronal targeting elements such as the synapsin promoter or microRNAs.

4. Expert opinion

Based on these developments in targeting the CNS, it is not hard to imagine a future where a patient goes into a hospital for a shot of a gene vector that would only specifically affect the targeted diseased cells and completely avoid the nontargeted cells, like a ‘magic bullet’ as envisaged by Paul Ehrlich for chemotherapy over a century ago. Specific pools of neuronal populations could be efficiently targeted throughout the nervous system without having to perturb the brain or spinal cord with a cannula. The field is working toward finer and finer pinpoint targeting of neurons, neuronal populations, and their subtypes [12]. However, our imagination may be beyond the ultimate precision of resolution and targeting capability necessary to affect or restore one CNS function.

Focal injections of AAV-synapsin promoter vector suggested a neuron-specific expression pattern [4] while intravenous delivery of an AAV9-synapsin promoter vector resulted in some degree of expression in peripheral tissues [8]. The lack of absolute neuron specificity could be due to the short recombinant promoter sequence or other factors. The AAV2 inverted terminal repeats contained within the construct confer weak promoter activity [13,14] which could be responsible for the non-neuronal expression seen with intravenous administration [8]. It also remains possible that the AAV2 terminal repeats could alter promoter activity and affect specificity. For example, AAV constructs containing an astroglial specific promoter were shown to express in neurons [15], so the fidelity of tissue-specific promoters may be compromised within the context of AAV. Nevertheless, it appears that AAV-synapsin promoter vectors will provide an indispensable approach to selectively target central neurons after a peripheral delivery. We expect more wide-scale use of this powerful and popular promoter in combination with other targeting elements including CRISPR gene editing.

Declaration of interest

We thank Thomas McCown and Xiao-Hong Lu for discussions. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.

Additional information

Funding

This study was supported by Amyotrophic Lateral Sclerosis Association; Fidelity Biosciences Research Initiative, and Karyopharm Therapeutics.