Drivers: Brain Circuits and Networks That Control Pain Behavior
In a keynote lecture, Howard Fields, University of California, San Francisco, US, took up the topic of what happens to the nociceptive signal once it reaches the brain. “Pain researchers have done a good job working out how the nociceptive signal is generated and processed,” said Fields. “But the more relevant biological question is how that nociceptive signal is used to guide behavior. We haven’t paid enough attention to that.” Fields now works mainly in the area of addiction and reward, and one important function of pain, he went on to explain, is how it impacts reward/decision-making circuits in the brain to affect behavior.
In 1927, Pavlov had discovered a critical feature of the interaction of pain and reward. In Pavlov’s classic experiment, a dog, conditioned to associate the ringing of a bell with a food reward, would eventually salivate at the sound of the bell. But in lesser known experiments, Pavlov gave hungry dogs an acute painful stimulus, followed immediately by feeding. He found that with time, the dogs no longer cringed and howled in response to the noxious cue, but instead began salivating and looking for food. How does that work?
That happens because responses to pain are not merely reflexive, but involve a decision-making process that weighs the costs and benefits of competing factors. For example, if an investigator places some cheese in a chamber where a rat has previously received a painful injection, the rat will need to decide if it is worth risking pain for the pleasure of a food reward. “That’s life in a nutshell,” said Fields.
Howard Fields
One essential site that mediates this decision process is the nucleus accumbens (NAc), part of the dopaminergic mesocorticolimbic reward pathway in the brain. The firing of NAc neurons predicts the value of anticipated rewards and drives appetitive behavior. The predictor neurons respond to dopamine released by projections from neurons in the ventral tegmental area (VTA). Aversive signals, including pain, also appear to be encoded in the NAc—human functional imaging studies show that pain predictive cues activate NAc regions that overlap with those activated by reward predictive cues (Jensen et al., 2003).
One way the brain may resolve the conflict between pain and pleasure is with endogenous opioids. In some classic experiments, investigators set rats on a hot plate (turned off) and gave them either regular lab chow or chocolate to eat. Then, the researchers turned on the hot plate—as the temperature rose, the animals eating lab chow jumped off after just a few seconds, but rats eating chocolate stayed on nearly twice as long. If the rats were given naloxone to block opioid signaling, they jumped off right away, whether they were eating regular food or chocolate (Dum and Herz, 1984). The hot plate served as a conflicting motivator: when off, it predicted the reward of yummy chocolate, but when turned on, it became a disincentive to keep eating. That conflict was resolved in part by the release of endogenous opioids.
This type of conflict resolution might also explain how placebos work. The placebo effect is blocked by naloxone (Amanzio and Benedetti, 1999), and placebo treatments are known to cause release of endogenous opioids in the NAc (Zubieta et al., 2005). Fields said this is consistent with placebo being a reward-predictive cue (the reward being pain relief), which causes endogenous opioid release in the NAc.
In the hot plate experiment, activity mediated by endogenous opioids in the NAc drives the decision to stay and eat the chocolate, but animals also need to suppress other actions (e.g., the desire to jump off the hot plate). This could be a role for descending pain inhibition, as a way to support the pursuit of pleasure. Eating, which activates NAc reward pathways, has also been shown to inhibit pain-facilitatory neurons (ON cells) and activate pain-inhibitory neurons (OFF cells) in the brainstem (Foo and Mason, 2005). Descending pain inhibition occurs via a serially connected circuit involving the hypothalamus, periaqueductal gray, rostral ventral medulla, and dorsal horn of the spinal cord. Opioids act at every level of this pathway. Field said that one reason these drugs work so well might be because they target both the sensory and the motivational aspects of pain.
What does all of this mean for chronic pain? The balance between pain and reward is struck outside of consciousness, before the animal or person even feels the sensation, Fields said. Pain, then, serves as a motivational force that can help organisms make better decisions on the fly. But chronic pain sets up a situation where the decision circuit may be or may become biased to responding to pain. “Chronic pain can be looked at as a motivation state that is going to bias you toward responding to pain as opposed to any other competing motivations,” Fields concluded. This new approach to understanding pain broadens the relevant CNS circuits and may open up opportunities for treatments geared at reward networks (for more on this, see a recent review by Navratilova and Porreca, 2014).
In the following session, Neil Schwartz, Stanford University, California, US, presented his work on how chronic pain causes a drop in motivation to work for rewards by eliciting maladaptive synaptic plasticity in the NAc in mice. His work was recently published (for complete coverage of Schwartz’s work, see PRF related news story on Schwartz et al., 2014).
Attention grabber
Karen Davis, Toronto Western Research Institute and University of Toronto, Canada, studies the interaction between pain and the brain circuits governing attention. There is a lot of variability among people in the connectivity and strength of brain networks (see Mueller et al., 2013), and Davis thinks that individual differences in networks involved in handling pain will open up opportunities for personalizing treatments for chronic pain.
Karen Davis
In a string of papers published in the last two years, Davis has demonstrated that people differ in their ability to pay attention to a task during a painful stimulus. She identified two types of people: those who are distracted by pain (P types) and those who can keep their attention on a task even in the presence of pain (A types). Davis and coworkers showed that P types and A types have different brain structure and network function (Erpelding and Davis, 2013; Seminowicz et al., 2004). More recently, the group reported that the strength and variability of connections between the periaqueductal gray region in the descending pain modulatory system and the default-mode network (a brain network active when the brain is at rest) correlate with how effectively people can ignore or disregard pain (see PRF related news story on Kucyi et al., 2013, and also Kucyi et al., 2014). The idea, Davis said, is that A types and mind-wandering types, who are better able to get their minds off their pain, might be more amenable to cognitive therapies to control pain, whereas P types may need other approaches.
In the question session, Davis was asked if she thinks the tendency to be an A versus a P type person is a stable trait. She said it seemed that way in the context of their studies, but the question is whether you can change it with cognitive behavioral therapy (CBT). “That depends on whether the brain is rigid or not,” Davis said. “The hope is that we can turn some of the P people into A people.”
While that prospect remains unclear, Davis has shown that people who do modify pain-related thoughts can exert control over centrally mediated pain sensitivity (Salomons et al., 2014). In a recent study, CBT resulted in a reduction in secondary hyperalgesia, a measure of central sensitization in healthy adults exposed to painful heat. In subjects receiving eight brief sessions of CBT, the area of secondary hyperalgesia was reduced compared to subjects receiving a non-pain-related psychological intervention. Davis said her group is now looking at the brain basis of this phenomenon.
Wiring diagrams
Katja Wiech, University of Oxford, UK, is looking to brain anatomy and function for clues to the wiring of pain circuits, and individual variation therein. She recently completed a study looking at how the insula, a key node in the brain matrix that responds to both acute and chronic pain, connects to pain circuits and whether differences in connection could 
Katja Wiech
parallel pain-relevant psychological traits. The insula is divided into three regions, each thought to play a different role in pain perception: the anterior insula is associated with the cognitive-affective responses to pain, while the middle and posterior parts are associated with sensory-discriminative aspects. Wiech’s new data looking at either structural connectivity using diffusion tensor imaging tractography or functional connectivity by resting-state fMRI in healthy adults were consistent with this idea: the anterior insula appeared connected predominantly to affective brain regions in the ventrolateral prefrontal cortex and orbitofrontal cortex, while the posterior region showed connections to sensory cortices. The mid-insula was connected to both sensory and affective regions. On an individual level, the patterns of connectivity correlated with pain-related traits: pain vigilance and awareness was positively correlated with connectivity between the anterior insula and the amygdala, and negatively correlated with connectivity with the rostral anterior cingulate cortex in the descending pain control pathway. The work was recently published (Wiech et al., 2014).
In a short talk, Jing Wang, New York University, US, presented new data using optogenetics to identify a novel brain circuit that regulates neuropathic pain. Optogenetic approaches have been used to activate or quiet primary sensory neurons (see PRF related news story on Iyer et al., 2014). In the new work, Wang and his colleagues looked at the role of neurons in the prefrontal cortex (PFC), an important brain region for top-down control of pain. PFC neurons project to the NAc, a key component of reward pathways, but the role of this connection is not known. It has been shown that reorganization of the PFC-NAc circuit predicts the transition from acute to chronic low back pain (Baliki et al., 2012).
In the study, Wang and colleagues expressed light-activated channelrhodopsin in prefrontal cortical neurons in rats and showed they could trigger action potentials in the neurons with blue light, both in ex vivo brain preparations and in vivo. The light treatment could completely reverse mechanical and cold allodynia after spared nerve injury in rats. Activation of the PFC neurons also eased anhedonic symptoms of pain. Injured rats lose their preference for the pleasing taste of sucrose, but their consumption returned to normal levels with light exposure. PFC activation had no effect on uninjured rats in measures of pain or sucrose preference.
In this model, Wang’s group could also look at activation in the projections of the PFC neurons by shining light on the terminals where they synapse with medium spiny neurons in the NAc core. The researchers showed that light directed at the NAc caused spiking of post-synaptic NAc neurons, which was blocked by the glutamate receptor inhibitor NBQX. Activation in the NAc reversed mechanical and cold allodynia in nerve-injured rats, although not as dramatically as PFC activation. The effects of light on pain, as well as sucrose consumption, are all blocked by NBQX. Wang calls this “a novel cortico-striatal pathway for pain control” that provides inhibition of both sensory and affective components of chronic pain.
For other coverage of the meeting, see Part 1 and Part 3.
Image: © iStockphoto.com/raisbeckfoto
Article Contributors:
