Live-cell imaging methods for the study of vagal afferents within the nucleus of the solitary tract

Abstract

Substantial evidence suggests that vagal afferent functions are modulated by agonists acting on afferent terminals in the solitary nucleus (NST). Actions of these agonists are implied through intracellular recordings from cultured nodose ganglion cells or second order NST neurons. While high-quality data have been obtained using these methods, techniques in which physiological measurements can be made directly on the afferent terminal fields, in situ, in the NST, would eliminate several potential interpretive problems inherent in these less direct approaches. This paper describes methods developed to directly measure changes in presynaptic cytoplasmic calcium in vagal afferents using time-lapse laser confocal microscopy on the in vitro brainstem slice. Calcium green dextran (CG) transported from the nodose ganglion clearly demonstrates vagal afferent fibers ramifying throughout the NST in the in vitro brainstem slice. CG-labeled vagal afferents can be repeatedly activated by focal electrical stimulation, by agonists acting on presynaptic ligand-gated ion channels, and by molecules that are presumed to act directly on vagal afferents based on previous physiological and immunocytochemical studies. Image and preparation stability are a challenge to the success of the experiment; however, methods described here should assist direct studies of transduction events within other afferent terminal fields in the CNS.

Introduction

The smooth, coordinated neural control of a wide range of homeostatic processes depends on general and special visceral afferent projections to the nucleus of the solitary tract (NST) in the caudal medulla (Gillis et al., 1989, Rogers et al., 1995, Rogers et al., 1999, Rogers et al., 2005). For example, digestive, especially gastric, functions are under the dominant control of brainstem “vago-vagal” reflexes. Parallel control of several aspects of ingestive behavior, such as the selection, timing, and quantity of individual meals, is also regulated by visceral afferent input to the NST (Grill et al., 2002). How the NST sorts this vast assembly of data to regulate virtually all aspects of bodily homeostasis is an intensely researched question (Blessing et al., 1997, Shapiro and Miselis, 1985, Grundy and Schemann, 2002, Altschuler et al., 1989, Rogers and McCann, 1993, Rogers et al., 1995, Rogers et al., 1999, Rogers et al., 2005, Goyal et al., 2001, Powley and Phillips, 2002, Chang et al., 2003). Clearly, factors that affect the communication between vagal afferents and the NST should have profound and widespread effects on homeostasis.

Interest in vago-vagal reflex circuits has increased due to observations that descending CNS inputs and circulating signal molecules released during changes in digestive status, activity, stress, or disease can radically alter gastrointestinal (GI) function via action on medullary reflex control circuits (Gillis et al., 1989, Rogers et al., 1995, Travagli and Rogers, 2001, Travagli et al., 2003, Rogers et al., 2005). This realization has accelerated mechanistic physiological and pharmacological studies on this circuitry over the last 10 years (Rogers et al., 2005). Most studies of hormonal and CNS pathway modulation of brainstem gastrointestinal (“vago-vagal”) reflex function have focused on the action of extrinsic hormones and transmitters to regulate the activity of the NST (sensory) and DMV and NA (motor) components of vago-vagal reflexes (Rogers et al., 1995, Rogers et al., 2005, Blessing et al., 1997, Travagli and Rogers, 2001, Travagli et al., 2003). Recently, however, it has become clear that some extrinsic signal molecules may act potently and, in some cases, exclusively on vagal afferent terminal fields within the solitary nucleus (Emch et al., 2001, Hoang and Hay, 2001, Doyle et al., 2002, Doyle et al., 2004, Hermann et al., 2004a, Jin et al., 2004, Page et al., 2005). These effects are reflected in significant changes in a variety of homeostatic processes (Hermann and Rogers, 1995, Emch et al., 2000, Page et al., 2005).

Modulation of vago-vagal reflexes may occur as a consequence of agonist action directly on central vagal afferents. For example, tumor necrosis factor (TNF) is a cytokine released by macrophages and activated microglia in response to antigenic challenge, radiation exposure, and injury (Dinarello, 1992, Vassalli, 1992). Its release is associated with the induction of fever, nausea, gastric stasis, emesis, and anorexia (Plata-Salaman, 2000). There is good evidence for vagal involvement in each of these effects. Immunocytochemical studies show that the TNF-type one receptor (p55 or TNFR-1) is constitutively expressed on vagal afferent fibers (Hermann et al., 2004a). Furthermore, the effects of TNF to activate NST neurons appear to depend on intact glutamate neurotransmission within the NST (Emch et al., 2001) and vagal afferents utilize glutamate as a neurotransmitter substance in the NST (Andresen and Yang, 1990, Smith et al., 1998, Yen et al., 1999, Hornby, 2001, Travagli and Rogers, 2001, Travagli et al., 2003). Indirect evidence suggests that a number of other agonists (e.g., glutamate, cholecytokinin (CCK), leptin, ATP, and capsaicin) could affect vago-vagal reflex control perhaps via direct action on vagal neurotransmission within the NST as well (Hoang and Hay, 2001, Simasko and Ritter, 2003, Jin et al., 2004, Peters et al., 2004, Dang et al., 2005, Page et al., 2005).

Agonist effects on central vagal afferents have been inferred by a combination of methods. Immunohistochemical demonstration of an agonist receptor on vagal afferents is certainly supportive of the hypothesis that the agonist modulates vagal neurotransmission. The mere presence of receptor-like immunoreactivity does not prove receptor function at the mechanistic level. Nevertheless, electrical recordings and calcium imaging techniques reveal that nodose neuron cell bodies do possess functional receptors for a number of potential signal molecules (Simasko and Ritter, 2003, Peters et al., 2004). Nodose recording studies have, thus far, provided the only direct physiological information available on agonist action on vagal afferent neurons (Simasko and Ritter, 2003, Peters et al., 2004). While valuable insights into vagal afferent sensitivities have been obtained using these methods, results obtained from these studies have been subject to criticisms, such as:

• the receptor population in the cell body may not accurately reflect the distribution of receptors in the afferent fiber or terminals;

• the nodose receptor population may not be exposed to the agonist under physiological conditions;

• the transduction machinery in the terminal may not be the same as in the cell body; or

• the extraction and culture methods used to prepare neurons may significantly alter receptor population and functionality (Scott and Edwards, 1980, Mateos et al., 1998, Alvarez et al., 2000, Lancaster et al., 2001, Hermann et al., 2004a).

Results from in vitro whole cell voltage and current clamp studies of NST neurons strongly imply that some agonists act directly on vagal afferents. The essential effects of vagal input to NST neurons studied using these methods (i.e., the generation of glutamatergic post-synaptic currents) are not controversial (Doyle et al., 2002, Doyle et al., 2004, Jin et al., 2004). On the other hand, observations of the modulation of that process can be open to interpretation given that agonists and drugs acting on receptors and transduction processes in the presynaptic terminal may also act on the recorded post-synaptic cell.

The criticisms of these methods may or may not be valid. Nevertheless, until a means is developed to directly examine presynaptic vagal activity, in situ, criticisms will remain. Direct electrical recordings from vagal fibers and terminals, whose cross-sections measure ∼1 μm maximum, are improbable with present methods (Boyer et al., 2004).

While other in vitro model systems, such as hippocampus, cerebellum, tectum, and cortex possess highly structured and laminar relationships between afferent inputs and neurons receiving those inputs, the vagal–NST relationship is diffuse (Doyle et al., 2004). This anatomical feature further complicates the physiological analysis of vagal afferent inputs to NST neurons. As a consequence, imaging and neurophysiological methods that are appropriate for study of en mass discharge properties of a whole field of well structured afferent terminals must be replaced by methods capable of imaging small clusters or even individual varicosities.

Fluorescent imaging of the response of live cells labeled with a calcium fluorophore may provide a direct means to verify the site of action of agonists. This reasoning is based on the fact that the dominant controlling factor for transmitter release from presynaptic terminals is cytoplasmic calcium (Fisher and Bourque, 2001, Fill and Copello, 2002). The presynaptic calcium pool linked to vesicle exocytosis is under the control of several factors, such as voltage- and ligand-gated calcium channels on the presynaptic membrane as well as intracellular calcium storage pools in the endoplasmic reticulum. The regulation of these sources of cytoplasmic calcium is largely responsible for the short-term modulation of synaptic transmission (Fisher and Bourque, 2001, Fill and Copello, 2002, Lelli et al., 2003). Therefore, visualization of terminal and preterminal cytoplasmic calcium signals in vagal afferents in the NST, in situ, could provide a powerful tool to investigate the transduction mechanisms by which agonists effect changes in vagal afferent synaptic transmission. Additionally, these calcium-imaging methods may be used in young adult rodents whereas whole-cell slice recording methods are usually restricted to neonates and juveniles due to the overgrowth of neuropil and glia that occurs with maturation.

Our initial use of live cell laser confocal microscopy was applied to the in vitro slice preparation to investigate the actions of agonists on identified neurons of the NST (Hermann et al., 2005). Our experience in imaging complex agonist-induced changes in cytoplasmic calcium in NST neurons (∼6–8 μm diameter) led us to conclude that we could image calcium changes in vagal afferents in the NST as well. Our confidence was reinforced by recent reports by Lelli et al. (2003) and Conti et al. (2004) showing that data relevant to the regulation of terminal calcium levels could be obtained in in vitro slice preparations. The methods described in this report outline the instrumentation, essential protocols, means of analysis, and pitfalls involved in performing these observations. While these methods focus on calcium imaging from vagal afferent terminals in the NST, they should be applicable with modification to the study of any terminal field within the CNS.