Article In Brief
In animal models of stroke and traumatic brain injury, researchers found that inhibiting activity levels of a protein kinase called CC chemokine receptor 5) led to improvements in motor and cognitive deficits. The research team is next going to study the effect in human trials.
A team of scientists at University of California, Los Angeles (UCLA) have discovered a new target involved in a molecular memory system—a protein kinase called CCR5 (CC chemokine receptor 5)—that increases in neurons on the heels of stroke and traumatic brain injury (TBI).
Knocking down levels of this protein in animal models of stroke led to a faster and stronger recovery of motor function. Similarly, reducing CCR5 following a TBI led to significant improvements in memory and learning, findings that suggest targeting CCR5 may be a novel treatment for brain recovery after stroke and TBI.
Once the scientists identified the target, known for its involvement in learning and memory and as a co-receptor for HIV, they set about to see whether people who carry a gene allele that silences CCR5 (making about 50 percent less of it) have better recovery after stroke. They found they did.
The culmination of their studies, published in the February 21 online issue of Cell, led the group to initiate a clinical trial in stroke patients. They are testing an HIV drug, maraviroc, that hits the target and reduces activity levels of the protein. The trial has just begun.
“Companies have been struggling to find a drug that taps into a universal recovery pathway for stroke and traumatic brain injury. We think we have found this pathway,” said the senior study author, S. Thomas Carmichael, MD, PhD, professor and Frances Stark endowed chair of neurology at the David Geffen School of Medicine at UCLA. “It is very exciting. We identified a fairly dramatic effect on brain plasticity and recovery.”
The story of this scientific journey began two years ago at UCLA when Alcino J. Silva, PhD, distinguished professor in the departments of neurobiology, psychiatry and biobehavioral sciences and psychology, and director of the Integrative Center for Learning and Memory at UCLA's Brain Research Institute, and his colleagues published a study in eLife that identified CCR5's role in learning and memory.
CCR5 sits on the cell surface and inhibits cAMP responsive element binding protein (CREB) and other signaling molecules that are critical for memory to take place. Before this, CCR5 was known to be involved with immune trafficking of neutrophils and T lymphocytes. It had not been identified as a molecule associated with brain function. This finding was then substantially accelerated in studies of brain disease when more than two dozen scientists, including Drs. Silva, Carmichael, and TBI scientist Esther Shohami, PhD, professor emeritus in pharmacology at the HU Institute for Drug Research in Jerusalem, came together to study the regeneration of brain, spinal cord, and peripheral nerves. The current paper is the culmination of their research.
Study Design, Findings
After the Silva lab published their study showing that knocking down CCR5 led to enhanced memory, the UCLA collaborators designed the study to look at CCR5 and its role in brain repair in animals following stroke. Indeed, they showed that stroke induces CCR5 in cortical neurons. Normally, CCR5 gene expression is undetectable in neurons. But after a stroke, there was a 100 million-fold induction in cortical neurons during the first hours and days post-stroke and the levels stayed high for months. There was evidence that these high levels were getting in the way of recovery.
The researchers asked: What happens if these high CCR5 levels are reduced in post-stroke neurons? Would animals recover faster, better? And if so, what was going on in the molecular pathway that allowed this recovery to take place?
To answer these questions, the scientists first paired a gene and viral vector to target the regions around the site of the stroke to reduce CCR5 in cortical neurons in motor to pre-motor cortex. They introduced the viral vector three days before the insult. (Some of the control animals received a viral vector without an active gene and others with just a stroke and no intervention were tested.) The vector was delivered anterior to the stroke site, in the part of the brain that normally has a limited recovery in stroke. The timing led to a knockdown of CCR5 in the animals that had received the active gene therapy. The decrease was observed within a week of the stroke.
The scientists measured functional recovery in two motor tasks, one testing forelimb activity in rearing and the other in gait. The motor deficits persisted throughout the nine weeks of testing but in animals that received the gene/viral vector, a knockdown of CCR5 resulted in dramatically fewer foot missteps, or faults: 5.9 percent for the experimental group versus 11.25 percent for controls in the grid walk test. The improvements persisted over the two-month study period. The animals also showed significantly better use of their forelimbs (as they moved around a transparent cylinder) compared with the control animals. Again, the benefits continued over the course of the experiment.
The researchers repeated the same studies using the HIV drug, maraviroc, a CCR5 antagonist that reduces levels of the protein. In 1996, CCR5 was identified as a co-receptor of the HIV virus, and this finding led to the development of the antagonist. If the drug worked in inhibiting CCR5 activity in cortical neurons, the scientists would have a ready-made treatment to test on patients. Again, they wanted to see whether it would enhance recovery—and it did. They delivered intraperitoneal injections of the drug (100 mg/kg) daily starting at 24-hours post-stroke through the nine weeks of the testing period. They used ultra-performance liquid chromatography to ensure the drug got into the brain in a dose sufficient to inhibit CCR5.
The treatment resulted in improved motor control on the grid walk but the benefits were not observed until three weeks post-stroke: 7.5 percent for the treated group versus 15.7 percent of missteps for the control arm. The improvements persisted.
The investigators conducted another study to see whether the benefit would continue if the medication stopped. They treated one group for the first three weeks and tested them at the end of the nine weeks. The others in this treatment arm continued to receive daily doses throughout the nine weeks. At three weeks, there were 29 percent missteps in the treated group compared with 39 percent foot faults in the controls. When the medication stopped, the animals continued to perform better in the grid walk task: 27.4 percent for the treated group compared with 37.3 percent for those who did not receive the treatment.
The treated animals that received no more medication after three weeks also showed sustained improvements on the cylinder test.
Next, the investigators set out to see whether they could push the recovery envelope on chronic stroke. The animals were treated with maraviroc at three weeks post-stroke and the motor deficits were measured up to 11 weeks post-stroke. Again, the treated animals had fewer foot faults at eight weeks and at 11 weeks compared to the two control groups: 28 percent versus 40 percent at eight weeks; and 24.7 percent versus 38 percent at 11 weeks. Forelimb improvements were also more common in the delayed treatment group.
Tests on brain tissue showed that treatments preserved dendritic spines (but did not grow new ones) and helped increase axonal sprouting and form new brain connections. The scientists believe that the knockdown of CCR5 produced motor recovery through CREB and DLK (dual leucine zipper kinase) signaling. These downstream targets increased following the treatments.
Focusing on TBI
The researchers also focused on TBI, collaborating with Dr. Shohami, who conducted all of the TBI studies in the paper. The team used the closed head injury model she developed and showed that CCR5 also increases in cortical neurons at the time of injury. They wanted to know whether knocking down CCR5 would lead to improvements in learning and memory. This time, the AAV vector was delivered to the CA1 and CA3 regions of the hippocampus two weeks before a closed head injury and evaluated on a series of memory tasks 48 hours later. Were they able to perform better than controls on novel object recognition? Yes, they were. (The experimental group spent more time with the “novel” object than with an object they were familiar with: 74 percent versus 59 percent in the control group, respectively.)
The mice were also evaluated in a Barnes maze test between seven to 10 days post-TBI. After the same amount of training as controls, the animals with reduced CCR5 showed significant improvements in latency and they were able to find the hole to the goal box much faster than controls. The treated animals performed almost as well as the healthy sham animals, according to the scientists.
They repeated the studies using maraviroc treatment after TBI and also observed significant improvements in both behavioral tests four days after the head injury.
From Animals to Humans
Many stroke treatment studies have looked great in animals and did not pan out in clinical trials. The UCLA investigators knew about a common loss of function CCR5 allele from the HIV literature. People who carry one copy of CCR5-Δ32 (about 8 to 10 percent in Northern European populations) seem to have protection against HIV infection. They have naturally occurring low levels of CCR5. The gene does not make the CCR5 protein so people make about half of the protein through the normal allele.
The question was this: If they were right, those who inherit this allele might fare better following a stroke. They knew that Ashkenazi Jews had higher rates of the CCR5-Δ32 allele and identified what they thought would be the perfect study to see whether patients had better post-stroke outcomes. They reached out to Natan Bornstein, MD, professor of neurology and head of the stroke unit at Tel Aviv Medical Center. He's been running the Tel Aviv Brain Acute Stroke Cohort, known as TABASCO, since 2008. They have amassed demographic, psychological, inflammatory, biochemical, neuroimaging and genetic markers data on hundreds of people after their first (mild to moderate) stroke. The hope is to identify predictors of recovery. Throughout the study, they measured cognitive deterioration, vascular events (including recurrent strokes,) falls, affective changes, functional difficulties and mortality. Dr. Bornstein was happy to collaborate.
The investigators pored through 446 patient records. Sixty-eight of these patients (15.2 percent) had a CCR5-Δ32 allele. These patients made significantly better recovery on a range of neurological studies (cognitive and functional outcomes) than those without this allele. They had better performance in memory, verbal function, attention and measures of overall cognition than patients than non-carriers. There were no differences in executive function and visuospatial measures. They also fared better on functional outcome measures used to assess stroke recovery.
“These results are consistent with our animal studies,” said Dr. Carmichael. The results from the Tel Aviv analysis led them to design a phase 2 study in humans. They want to find patients between one and six weeks post-stroke, which is a challenge since many of them are in private rehabilitation settings. They have just started recruitment.
“These findings are very exciting and encouraging,” added Dr. Silva. “We are on the verge of developing a brand new treatment for stroke. We may be able to come in with rehabilitation and drug treatments that enhance plasticity. It would be amazing.”
Commenting on the paper, Jin-Moo Lee, MD, PhD, the Norman J. Stupp professor of neurology and the head of the cerebrovascular disease section at Washington University School of Medicine & Barnes-Jewish Hospital, said: “This is a landmark study that identifies a drug target, examines its mechanism of action, and demonstrates efficacy in an animal model of stroke recovery. Moreover, the relevance of this drug target in human stroke is convincingly shown using genetics.”
“Previous stroke studies using animal models have not always translated into successful human trials,” Dr. Lee added. “But in this study, the authors demonstrated that individuals with a genetic variant in the CCR5 gene influences recovery after stroke.”
A drug targeting CCR5 has already been approved by the US Food and Drug Administration (FDA) for use in AIDs prevention, Dr. Lee noted. “This drug can now be repurposed to enhance brain repair after stroke. The science in this paper is rigorous and very exciting. If this drug proves to be efficacious in human trials, it would be a game changer.”
Rajiv R. Ratan, MD, PhD, executive director of the Burke Neurological Institute at Weill Cornell Medicine, agreed. “Part of the major challenge in developing a therapy for stroke recovery is there hasn't been an adequate understanding of the molecular targets. This group not only identified a target but they demonstrated that knocking down the protein leads to a robust modulation of recovery after stroke and traumatic brain injury.
“This tour de force provided genetic data from humans that reinforces the idea that lowering CCR5 will be a good therapeutic strategy for stroke patients. It is a remarkable study on many levels.
“Obviously, great research raises other exciting questions and opportunities. It will be important to have a biomarker for identifying the right dose of the CCR5 antagonist as these studies move into humans with stroke or traumatic brain injury. Also, one could imagine additional mechanisms are involved and it would be good to clearly establish what is causing the increase in contralateral connectivity to the damage area around the stroke, and whether this is a critical component of the recovery that they measured.”
Dr. Carmichael's laboratory has received grant funding from Takeda Pharmaceuticals. Drs. Lee and Ratan had no disclosures.