I am a
Home I AM A Search Login

The Missing Link Between Disinhibition and Excitation in the Spinal Cord During Chronic Pain: Taking a STEP Forward

Converging evidence from rodent and human tissue points to the phosphatase STEP61 as a hub of dorsal horn pain processing

by Neil Andrews


15 July 2019


PRF News

STEP61Featured

Converging evidence from rodent and human tissue points to the phosphatase STEP61 as a hub of dorsal horn pain processing

Spinal cord pain processing is a complicated phenomenon involving different neurobiological mechanisms and molecular players. Dorsal horn disinhibition spurred by brain-derived neurotrophic factor (BDNF) and excitation mediated by NMDA receptors (NMDARs) have been implicated in this complex activity, but the molecular link between the two remained unknown.

 

Now, new research from Michael Hildebrand, Carleton University, Ottawa, Canada, and collaborators, using electrophysiological and biochemical techniques, identifies STEP61, a protein tyrosine phosphatase, as the bridge between disinhibition and excitation in the spinal cord, serving as a molecular brake that prevents pain. This conclusion stems from evidence not only from rodent models of inflammatory pain but also from a human pain model. Most of all, the work shows that it’s possible to use tissue from the human spinal cord to investigate pain mechanisms and determine if what happens in animal models applies to people too.

 

“The most profound long-term implication of this paper is that it shows the feasibility of using human spinal cord tissue for both electrophysiology and biochemistry,” said Michael Gold, University of Pittsburgh, US. “I don’t know of anyone who has recorded from human spinal cord slices before. To do something like that is huge because of the potential to understand the unique features of human dorsal horn physiology relative to the rodent,” according to Gold, who was not involved in the research.

 

The study appeared June 1, 2019, in Brain.

 

What holds true for neuropathic pain also holds true for inflammatory pain

Previous work by Michael Salter, current study co-author Yves De Koninck, and colleagues set a framework for understanding spinal cord pain processing that undergirds the new study (see PRF related news stories here and here). The essential idea is that upon traumatic nerve injury, microglia in the spinal cord become responsive to ATP via activation by P2X4, a receptor on these non-neuronal cells. The microglia then release BDNF, which acts on dorsal horn neurons to decrease activity of KCC2, a potassium-chloride co-transporter. The end result is an increase in intracellular chloride concentration and a loss of inhibition in the spinal cord thought to underlie hypersensitivity during neuropathic pain.

 

The current study picks up on that earlier story, aiming to identify and explain a link between disinhibition by KCC2 and excitation via GluN2B subunit-containing NMDARs, the latter a spinal cord mechanism also identified by Salter and colleagues. In previous work, Salter, Hildebrand, and colleagues had shown a direct connection between these two mechanisms following nerve injury (Hildebrand et al., 2016). In the new research, they wanted to see if that also held true for inflammatory pain, but also identify the molecular linker driving the coupling between disinhibition and excitation in the spinal cord, which was not yet known.

 

The researchers first looked to the complete Freund’s adjuvant (CFA) model of inflammatory pain. Electrophysiological recordings revealed a greater amplitude of NMDAR miniature excitatory postsynaptic currents (mEPSCs) in lamina I neurons of the dorsal horn in CFA rats, compared to naïve animals. Next, looking at superficial dorsal horn synaptosomes (the cell fraction of synaptic proteins from isolated nerve terminals), they saw potentiation of GluN2B-containing NMDARs in CFA rats, according to biochemistry analysis showing an increase in activated Fyn (a kinase enzyme that phosphorylates GluN2B-containing NMDARs) and in total and phosphorylated GluN2B. They also observed decreased KCC2 in the synaptosomes. These results confirmed a link between disinhibition and excitation.

 

“The first big takeaway was that this mechanism looks like it’s conserved; it’s present in the CFA model,” said Hildebrand.

 

Along similar lines, biochemistry studies in a rat ex vivo BDNF model in which the researchers applied recombinant BDNF to spinal sections from naïve rats also showed downregulation of KCC2 as well as upregulation of phosphorylated Fyn and total and phosphorylated GluN2B. This, too, supported the importance of the BDNF-KCC2-NMDAR pathway on the road from disinhibition to excitation.

 

With regard to their ex vivo BDNF model, “what stands out to me is its broad range of application,” said co-first author Annemarie Dedek. “The fact that we can have a model we can use on naïve tissue, and then apply something that mirrored the effects we saw in both the inflammatory pain model and in our previous neuropathic pain model, is just so powerful and interesting.”

 

But while the results from the CFA and BDNF animal models were similar to the researchers’ earlier findings in their nerve injury model, the molecular linker coupling disinhibition to excitation remained elusive…

 

Solving a mystery

To identify the linker, the researchers set their sights on STEP61, since previous work from co-author Paul Lombroso, co-first author Jian Xu, both from Yale University School of Medicine, New Haven, US, and collaborators had found connections between STEP61, BDNF, and NMDARs (Xu et al., 2015Saavedra et al., 2016). There was also evidence to suggest that the phosphatase was involved in pain.

 

“It looked like STEP61 could be a good candidate in pain. We reached out to Paul and Jian, and that’s how this part of the research started, in terms of using their tools and really working together as a team to try to think about this mechanism and probe it,” Hildebrand told PRF.

 

It turned out to be a good approach. Indeed, the team saw a decrease in active STEP61 in superficial dorsal horn synaptosomes in both the in vivo CFA and ex vivo BDNF models. Further, by treating spinal sections from naïve rats with recombinant BDNF along with a fusion peptide enabling delivery of active STEP peptide into the cell, the researchers could prevent potentiation of NMDAR mEPSCs by BDNF, thus revealing a crucial role for STEP61 downregulation in that potentiation. Further implicating STEP61 were the results of experiments with a STEP inhibitor. When the researchers pretreated spinal sections from naïve rats with the inhibitor, BDNF now increased NMDAR responses. Biochemical experiments would further reveal loss of STEP61 as a key link between KCC2 downregulation and trafficking of NMDARs to synapses.

 

 

“When we restored STEP activity within the BDNF model, we were able to reverse the effect of BDNF, whereas if you drive STEP activity down, you can prime that effect,” said Hildebrand.

 

The group also used the carbonic acid anhydrase inhibitor acetazolamide, which blocks disinhibition produced by KCC2-dependent chloride dysregulation, in their ex vivo BDNF model. Co-treatment with the inhibitor and BDNF prevented the decrease in STEP61 that occurred in the synaptosomes of sections treated with BDNF alone. At the behavioral level, acetazolamide reversed the increased pain sensitivity (as measured by paw withdrawal thresholds) in CFA rats compared to CFA animals that received saline.

 

All in all, the electrophysiological and biochemical evidence showed an important role of STEP61 in the pathway leading from KCC2-dependent disinhibition to NMDAR potentiation.

 

Human tissue

With regard to their rat ex vivo BDNF model, Dedek said that “not only did it allow us to have a less invasive way to study [the BDNF-KCC2-NMDAR pathway], but it also opened the door to then use that model on human tissue as well.”

 

Indeed, it was a human ex vivo BDNF model to which the researchers next turned. To obtain the tissue, in this case from adult males, the researchers collaborated with co-author Eve Tsai, a neurosurgeon at The Ottawa Hospital and the University of Ottawa, Canada.

 

“The way it works,” said Dedek, “is that once consent has been obtained for organ donors to donate their spinal cord tissue for research, this tissue is recovered by a surgical team after the organs for transplantation have been recovered. The spinal cord is immediately placed in a protective solution to maximize tissue viability.“

 

The group first confirmed that lamina I neurons from the human tissue were viable and showed robust synaptic NMDAR responses to which GluN2B-containing receptors contributed. Then, they pretreated adjacent areas of lumbar spinal tissue with recombinant BDNF or saline, and then flash-froze the tissue. They discovered that human superficial dorsal horn synaptosomes from tissue treated with BDNF had decreased levels of KCC2 and total and active STEP61 protein, as well as increases in phosphorylated GluN2B and active Fyn. Using immunostaining, De Koninck’s research team also saw a significant decrease in KCC2 at neuronal membranes, compared to saline-treated controls, along with an increase in intracellular KCC2.

 

In short, evidence from rodent tissue and human tissue all pointed in the same direction: STEP61 serves as a molecular player linking disinhibition to excitation in the spinal cord. The researchers thus propose STEP61 as a molecular hub for spinal pain processing.

 

“Using different techniques and getting the same results makes for a compelling conclusion,” said Gold.

 

Open questions

Interestingly, and regardless of how the STEP61 story unfolds, the study doesn’t seem to quite fit, at least not yet, with a broader perspective in the pain field about inhibition/excitation in the spinal cord, according to Gold.

 

“The model the paper poses is a little bit perplexing to the extent that the loss of inhibition alone should be sufficient to produce hypersensitivity, based on everything we think we know about inhibition and excitation in the dorsal horn,” Gold explained. “So this work adds a level of complexity because what the authors seem to imply is that the sensitization one observes in the context of loss of inhibition isn’t due to the loss of inhibition per se, but to an increase in excitation, which is counterintuitive. The extent to which disinhibition alone is sufficient to mediate hypersensitivity, or whether you really need to take that next step and go from disinhibition leading to increased excitation will require further investigation.”

 

Another open question is whether STEP61 itself could be a good therapeutic target. Xu says that’s uncertain in the setting of chronic pain.

 

“STEP61 has been implicated in several other neurological disorders, including Parkinson’s disease, schizophrenia, and fragile X syndrome. In all those disorders, we’ve found an elevation of either STEP61 protein levels and/or activity, and we developed a small molecule inhibitor to target STEP61, showing a reversal of behavioral abnormalities in neurological models.”

 

“But in the pain models,” Xu continued “we found loss of STEP61 expression levels and/or activity, in which case we would need an activator to enhance STEP61 function. From a chemistry perspective, there aren’t many studies on activators or enhancers for a phosphatase. It might be more practical to do something like gene transfer using viral vectors to overexpress STEP61, but I’m not quite sure about this.”

 

On this issue, Gold said that “a number of different STEPs have been implicated in nociceptive pathways. Whether or not STEP61 is the be-all and end-all is not clear, since there are more STEPs out there that could complicate things from a therapeutic perspective.”

 

Regardless, Hildebrand says that the new work shows “the need for testing these mechanisms and pathways in human tissue models. Whether that’s using approaches like ours or genetic approaches from human chronic pain populations, I think that’s an urgent direction that’s needed, and our study helps start to address those kinds of questions.”

 

Neil Andrews is a science journalist and the executive editor of PRF.

 

Image courtesy of Dedek et al., 2019.

Share This