Chronic pain patients exhibit learning and emotional changes not often reversed by painkillers, including deficits in working memory, emotional decision-making, and classical conditioning tasks, as well as elevated anxiety and depression. Despite this knowledge, the role of the hippocampus, a crucial limbic area involved in learning and memory, in chronic pain has been largely unexplored. Now A. Vania Apkarian and colleagues at Northwestern University, Chicago, US, report that mice with nerve injury (a model of neuropathic pain) show defects in hippocampal-mediated behaviors, short-term synaptic plasticity, and the production of new neurons. First author Amelia Mutso and colleagues also document reductions in hippocampal volume in chronic pain patients that may, in part, correlate with the behavioral, cellular, and molecular abnormalities observed in the mice. The findings provide a new look at how chronic pain impacts the hippocampus, and could help explain deficits in emotion, learning, and memory seen in chronic pain patients. The findings appear in the April 25 Journal of Neuroscience.
“Pain perception has long been suspected to involve the hippocampus but our study now identifies the specific hippocampal processes and abnormalities involved in chronic pain,” Apkarian told PRF.
Recent work in people and using animal models of neuropathic pain has revealed that chronic pain is associated with a gross reorganization in cortical connections involving somatosensory, insular, and cingulate areas, as well as frontal and parietal regions of the brain. Although the hippocampus is not typically considered part of this “pain matrix,” the authors wanted to examine hippocampal function in chronic pain because the hippocampus plays a critical role in learning and memory, and because of recent studies that identified hippocampal abnormalities in animal models of chronic pain.
Apkarian and coworkers looked at hippocampal-dependent behavioral and cellular processes in mice with persistent allodynia after spared nerve injury (SNI), a model of neuropathic pain. In fear-extinction tests, the animals maintained tone-dependent freezing or contextual fear behaviors in response to foot shock experiments longer than did sham mice. In contrast, both SNI and sham animals successfully exhibited amygdala-mediated cue fear extinction. The results suggest a specific hippocampal fear conditioning defect in the SNI mice, rather than an abnormality in the broader fear conditioning circuitry. The SNI mice also exhibited increased hippocampal-dependent anxiety-like behavior.
In the hippocampus, phosphorylation of the kinases Erk1 and Erk2 have been shown to be critical for contextual fear extinction behaviors. Consistent with the behavioral studies. Apkarian and colleagues observed a decrease in Erk1/2 phosphorylation levels and Erk2 gene expression levels in the hippocampus of SNI mice. Along with the phosphorylation changes, they found a concomitant deficit in Erk-mediated short-term synaptic plasticity. Since these changes in Erk expression and signaling were distinct from those seen in studies of anxiety and depression, the investigators concluded that chronic pain, rather than general stress, leads to the observed deficits in contextual fear extinction and synaptic plasticity.
The birth of new neurons in the dentate gyrus of the adult hippocampus (neurogenesis) was previously reported to be involved in learning, memory, depression, anxiety, and impairment in contextual fear conditioning. When the researchers measured the levels of neurogenesis in the SNI mice, they found significantly fewer newborn cells as measured by double-labeling for neuronal markers and DNA replication. The researchers, therefore, hypothesize that the molecular and cellular changes in the hippocampus could account for the learning and behavioral defects observed in the SNI mice. They also note that, while decreased neurogenesis has been previously associated with depression, this is the first study linking it to chronic pain.
Changes in hippocampal function could be related to the production of cytokines in hippocampal tissue after nerve injury. Xian-Guo Liu and colleagues at Sun Yat-sen University, Guangzhou, China, recently showed that overproduction of TNF-α in the hippocampus of SNI mice and rats led to neuropathic pain, hippocampal dysfunction, and memory deficits, and that inhibition of TNF-α prevented these memory deficits (Ren et al., 2011). “Our latest unpublished work shows that treatment of mice hippocampal and spinal cord slices with TNF-α leads to different results in these two regions: a substantial decrease in the numbers of dendrites and dendritic spines in the hippocampus and a significant increase in the spinal cord. We haven’t yet looked at the effects on neurogenesis in these two regions, but we hope to do so in light of the Apkarian study,” Liu told PRF. Since SNI upregulates TNF-α, Liu hypothesizes that a decrease in hippocampal neurogenesis may also result from overproduction of TNF-α in the CNS.
To extend the study from the SNI mice to chronic pain patients, Apkarian and colleagues measured hippocampal volumes in subjects with chronic back pain (CBP), complex regional pain syndrome (CRPS), or osteoarthritis (OA). They found that patients with CBP or CRPS, but not those with OA, had significantly reduced bilateral hippocampal volumes. While it is unclear if hippocampal volume changes are a cause or an effect of chronic pain, or both, Apkarian and colleagues hypothesize that, just as in the SNI mice, altered Erk2 signaling and reduced neurogenesis in the hippocampus may also characterize chronic neuropathic pain in humans. If so, such changes could lead to alterations in synaptic plasticity, deficits in contextual fear extinction, and decreases in hippocampal volume. The authors are currently studying the functional and anatomical changes in the hippocampus of chronic pain patients with the hope of illuminating the mechanisms underlying hippocampal volume changes.
Raji Edayathumangalam is an instructor in Neurology at Harvard Medical School, Boston, US, and a freelance science writer.


Comments
David Borsook, Boston Childrens Hospital
Another Brain Region in the Complex Story of a Complex Disease
The paper by Mutso et al. (2012) contributes to an ever-increasing understanding that pain affects multiple brain systems. Indeed the term “pain matrix,” which never really held any rational definition, is further diminished in its meaning since brain systems are inherently so interconnected that it is now more and more difficult to exclude any brain region from pain processing (acute or chronic). Such is the case for the hippocampus as described by the group from Chicago who continue to publish new and exciting research.
Specifically, behavioral measures and the molecular changes are reported in in the hippocampus in a spared nerve injury (SNI) model of chronic pain in mice. Data demonstrating altered neurogenesis in the hippocampus is provided. Some of the behaviors evaluated (i.e., anxiety) may be more reflective of amygdala function as this region is well known to be involved in anxiety disorders such as PTSD and perhaps this is one reason that they did not show diminished cue extinction as seen in studies of anxiety disorders.
In parallel to the animal studies, volumetric evaluation of hippocampal changes in patients with chronic pain-induced changes in gray matter volume are presented. The observed animal changes may provide an underlying basis for potential changes in the hippocampus in human studies. The data provided on the preclinical study is robust and novel. The integration of behavioral and molecular changes focusing on the hippocampus in a chronic pain model has significant implications for understanding the human condition. The added data, showing alterations in hippocampal volume I patients provides intriguing translational connections – but clearly more work will be done to further understand these processes in chronic pain in humans.
As the authors note, the work adds to volumetric changes previously published for chronic pain in humans (Zimmermann et al., 2009). However, the current paper brilliantly notes a further consideration: that chronic pain is associated with “conditioning and extinction, and viewed as a state of continual learning through which aversive emotional associations are constantly made with otherwise incidental events due to the persistent presence of a patient’s pain.” One of the strange enigmas related to chronic pain, however, is that it is hard to remember or recapitulate in our minds like some other sensory experiences.
An alternative interpretation is that chronic pain is also involved in the stress response. The hippocampus is clearly involved in the stress response (Leuner and Shors, 2012; Frodl and O’Keane, 2012). Chronic pain is an ongoing stressor that leads to an alteration in allostatic load (McEwen and Gianaros, 2011; Borsook et al., 2012) that may contribute to altered changes through direct neuronal or indirect cytokine activity with the hippocampus. More research from this and other laboratories will continue to demonstrate the importance of hippocampal function in a condition where we have begun to better understand its central processing.
References
Borsook D, Maleki N, Becerra L, McEwen B. Understanding migraine through the lens of maladaptive stress responses: a model disease of allostatic load.Neuron. 2012 Jan 26;73(2):219-34.
Frodl T, O'Keane V.How does the brain deal with cumulative stress? A reviewwith focus on developmental stress, HPA axis function and hippocampal structure in humans. Neurobiol Dis. 2012 Mar 9. [Epub ahead of print]
Leuner B, Shors TJ. Stress, anxiety, and dendritic spines: What are the connections?Neuroscience. 2012 Apr 20. [Epub ahead of print]
Maras PM, Baram TZ Sculpting the hippocampusfrom within: stress, spines, and CRH.Trends Neurosci. 2012 May;35(5):315-24
McEwenBS, Gianaros PJ.Stress- and allostasis-induced brain plasticity.Annu Rev Med. 2011;62:431-45
MutsoAA, Radzicki D, Baliki MN, Huang L, Banisadr G, Centeno MV, Radulovic J, Martina M, Miller RJ, Apkarian AV. Abnormalities in hippocampal functioning with persistent pain. J Neurosci. 2012 Apr 25;32(17):5747-56.
Zimmerman ME, Pan JW, Hetherington HP, Lipton ML, Baigi K, Lipton RB. Hippocampal correlates of pain in healthy elderly adults: a pilot study. Neurology. 2009 Nov 10;73(19):1567-70.
John Quintner, Fremantle Hospital
David, I agree that the
David, I agree that the concept of a specific "pain matrix" cannot be supported by the evidence. Would you agree that "pain processing" is another such concept that we can now abandon? How can the brain "process" its own output, which we regard as being a lived experience?
Stephani Sutherland, www.stephanisutherland.com
I wrote about this paper for
I wrote about this paper for Scientific American MIND:
http://www.scientificamerican.com/article.cfm?id=how-chronic-pain-affect...