It is known that estrogen can mitigate neurological injury. However, to optimize estrogen’s benefits, it must be administered as soon as possible postinjury, preferably within the “golden hour.” We have found that the best means to deliver estrogen is in a water-soluble form, given intravenously or intraosseously. Using a lateral fluid percussion model of traumatic brain injury (TBI) with rats, we demonstrate lessening of their injury when treated with soluble estradiol or ethinyl estradiol-3-sulfate. Magnetic resonance imaging reveals reduction of cerebral edema and anisotropy, while invasive measurements demonstrate reduced intracranial pressure, improved cerebral perfusion pressure, and partial brain oxygen pressure. Longer term benefits in cognition and memory have been shown with improved Morris Water Maze performance. Our promising results with TBI (and previously severe hemorrhage) encourage us to pursue application of ethinyl estradiol-3-sulfate for human use.
KeywordsEstradiol, Estrogen, Ethinyl estradiol-3-sulfate, Golden hour, Intracranial pressure, Lateral fluid percussion, Magnetic resonance imaging (MRI), Morris Water Maze, Neuroprotection, Traumatic brain injury (TBI)
We wish to acknowledge the expert operators of the LFP model system and TBI assays, Drs. Ahmar Ayub, Betul Cam, Huadong Zeng, and Thian Ng. Thanks go to Dr. Candace Floyd of the UAB Physical Medicine and Rehabilitation Department for providing the neuronal stretch assay data, to Dr. Thomas VanGroen and the UAB Rodent Behavioral Assessment Core for Morris Water Maze studies, and to Dr. Kurt Zinn and the staff of the UAB Small Animal Imaging Shared Facility for MRI and PET-CT imaging data. This work was supported by a Department of Defense grant PTO75653.
The serious consequences of traumatic brain injury (TBI) have been naively minimized or suppressed for years. Not long ago it was easy to cast TBI as nothing serious, when delivering a “knockout” is the ultimate goal for a boxer, or a “good hit” marks a talented defensive back in American football. Indeed, TBI was euphemistically described as “shell shock” in the First World War and was often a pejorative for the victim’s character. However, more recently awareness in two areas has dramatically changed public perceptions for the consequences of TBI.
First, the exposure of our troops to the asymmetric warfare in the Middle East, with widespread use of improvised explosive devices (IED) has spotlighted TBI. These simple, inexpensive, yet brutally effective weapons have dramatically increased warfighter’s TBI cases ( ). IED explosions have caused the greatest number of injuries in Iraq and Afghanistan as compared to other large-scale conflicts ( ). This weapon’s blast wave-initiated brain damage (ie, closed head) is the most common source of TBI induction, where pressures can approach 1000 atm, with damage at a microscopic or subcellular level ( ). The extent of blast injury has not only raised TBI awareness but has also sent many TBI patients home to the United States. Estimates are that between 2001 and 2007, 320,000 returning veterans from Iraq and Afghanistan were TBI positive ( ). Sadly, while brain-injured warfighter’s numbers have expanded enormously, there are no effective pharmacological treatments for TBI, despite the fact that a great deal of effort has been expended examining various therapeutic agents for TBI, including progesterone ( ). Putting it bluntly, stated: “To date, there is no specific drug treatment for acute brain injury, and many seemingly promising agents emerging from pre-clinical animal models have failed in clinical trials.”
The second area promoting TBI awareness is sports-related injuries, which focus our attention when children or celebrity athletes are involved. While military TBI is most likely to be induced by blast-wave forces ( ), civilian TBI is more often the result of blunt trauma and concussion ( ). Despite different inductions, there is likely to be a substantial overlap for military and civilian TBI ( ), where both can trigger downstream neurodegenerative disease, increasing with injury severity. However, severity may be a misleading criterion per se. While the military injuries from blast waves are often classified as mild TBI, this “mildness” blurs with repetition of the insults ( ). The same can be said with sports-related mild TBI, where repeated blows to the head often lead to serious consequences, such as chronic traumatic encephalopathy or Parkinson’s disease ( ), sometimes arising years later.
The Golden Hour
Effective therapeutic treatments for trauma must be administered as soon as possible postinjury. This has promoted the “golden hour” concept in trauma care, a window of opportunity to optimally deliver life-saving treatment. The golden hour rule has been clearly validated for TBI by the Resuscitation Outcomes Consortium ( ). It was found that:
Among out-of-hospital trauma patients meeting physiologic criteria for shock and traumatic brain injury, there was no association between time and outcome. However, the subgroup of shock patients requiring early critical resources and arriving after 60 minutes had higher mortality.
One means to provide early TBI treatment is to enable delivery by first responders. Thus the treatment should be highly portable, space efficient, stable, and convenient to administer.
Mechanisms of Estrogen Action
Estrogen has multiple modes of action, which can be influenced by route of delivery and solubility of the estrogen hormone. It is well known that estrogen acts via nuclear receptors, where estrogen ligands entering the cytoplasm engage estrogen receptors (ER), inducing dimerization and migration into the nucleus, whereupon ER activate a variety of genes via genomic estrogen response elements ( ). There are two ER isoforms, ERα and ERβ, which are expressed with varying levels and tissue specificities throughout the body, for both females and males. A third ER exists in the form of a G protein-coupled receptor, initially named GPR30, but now called GPER. GPER is bound to the endoplasmic reticulum and initiates rapid “nongenomic” actions ( ). GPER also has the ability to mitigate experimental TBI ( ).
ER also exists within a complex of signaling proteins (ie, signalosome) on the plasma membrane. When engaged, these receptors are major effectors of the rapid, nongenomic response ( ). We feel that this latter response is most likely the initial pattern for exogenously administered E2, since physiological benefits are seen almost immediately after introduction of E2 ( ). Through the use of single photon emission computed tomography-computerized tomography (SPECT-CT), we have shown that estrogen enhances cardiovascular performance by increasing cardiac output and kidney and liver perfusion in the face of severe hemorrhage ( ). Interestingly, the level of blood in the brain was not increased for E2 treatment over vehicle controls. Perhaps this is not such a paradox, since the rats are anesthetized to keep them still for imaging, reducing the brain’s demand for blood. In addition, trauma-hemorrhage (T-H) rats are in shock after 60% blood loss, which further reduces brain metabolism. It should be noted that in all our experiments, E2 is administered an hour or more after the injury, be it TBI or T-H.
The Neuroprotective Effects of Estrogens
There are several published accounts that document the neuroprotective properties of E2 ( ), which suggest potential for therapeutic use. Indeed, E2 is intrinsic to the brain, with local production in both males and females. This brain-endogenous E2 is neuroprotective as well ( ).
Engler-Chiurazzi and his colleagues have published reviews detailing the breadth of E2’s benefits for cognitive aging and brain injury ( ), including a historical review of two decades of research specifically for Alzheimer’s disease and stroke ( ). Of particular interest in this latter paper is findings from a dissection of the estrogen molecule to learn what is essential to confer neuroprotection, using a large library of estrogenic derivatives. Substitutions on the steroid A ring of alkyl groups at the 2 or 4 carbon positions exhibited enhanced neuroprotection from cerebral ischemia, which was brain-specific, acting without stimulating peripheral tissues ( ).
Estrogen replacement therapy’s risk/benefit profile in postmenopausal women remains controversial, especially in light of the higher incidence of stroke in these women ( ). These authors found that the hitherto unexplored effects of E2 on astrocytes suggests promise for neuroprotection based on a murine model, which mimics estrogen replacement therapy in ovariectomized mice. This effect was found to be mediated via ERβ of astrocytes ( ). There is also a crucial role for astrocytes in stroke ( ), where astrocytes respond to ischemia with reactive gliosis, excitotoxicity, and neuroinflammation.
The brain is unique among organs because, owing to the blood–brain barrier, it is essentially a “closed” system where estrogen can be synthesized and used locally. For example, it has been found that hippocampus-produced E2 maintains long-term potentiation and thus synapse health in females, but not males ( ).
A fascinating example of the neuroprotective actions of endogenous gonadal estrogens has been explored by Azcoitia et al. in a neurotoxicity model ( ). If the excitotoxin kainic acid is injected systemically at low doses, it can cause hippocampal hilar neurons to degenerate. However, injecting kainic acid in female animals at the beginning of estrus (ie, maximum systemic estrogen level) produced no neuronal damage, while injection at proestrus (lowest estrogen level) caused loss of hilar neurons, as did injection into ovariectomized females. Similarly, intact males suffered no damage from kainic acid, while castrated males lost hippocampal neurons. This seeming paradox may reflect the conversion of testosterone into estrogen by aromatase in vivo. The importance of early delivery was noted. When E2 was injected concurrently with kainic acid into ovariectomized females, neuronal damage was prevented, but with a delay of 24 h, there was loss of hippocampal neurons ( ). Again exploiting a kainic acid toxicity model, it was found that selective estrogen receptor modulators (SERM) exhibit some similar benefits, but overall E2 was superior. For example, E2 substantially downregulated active gliosis, which was less efficacious with SERMs ( ).
Neuronal Stretch Assay
The neuronal stretch assay is an excellent in vitro correlate for TBI mechanics, as the injured brain experiences shear, stretch, and compression. Furthermore, the assay reduces use of rats by deriving multiple assays from a small number of rat pups. This assay relies on the growth of explanted newborn rat neurons in tissue culture, grown on a flexible silastic membrane. This apparatus precisely and reproducibly deforms the membrane momentarily, producing the “stretch” injury.
Multiple Therapies from One Drug
In light of the foregoing, three significant points are noteworthy. First, we can extend life in animals from otherwise universally fatal 60% hemorrhage to 6 h with E2 and no other intervention, including resuscitation, suggesting that early E2 administration may extend the golden hour. This could be of great value to treat TBI from the battlefield or remote civilian areas. Second, if E2 is administered as a therapeutic drug for obvious severe hemorrhage as part of a polytrauma cluster, it should likewise be treating the pathologies of TBI in a “piggyback” fashion, where TBI may occur concurrently with blood loss and other injuries. Third, while we will not discuss sepsis here, we nonetheless have a basis to suggest that administration of E2 may mitigate the threat of sepsis in injured patients. This has been demonstrated in rat models which mimic penetrating wounds with necrosis, where the cecum is perforated to release gut flora into the peritoneal cavity, and ligated to induce necrosis ( ). In our studies, E2 administration after hemorrhage was followed by sepsis induction by cecal ligation and puncture. E2 produced improved immune function, cardiac output, splanchnic perfusion, and oxygen utilization ( ). A diagram of the possible “triple therapy” is seen in Fig. 10.7 .
Formulating a Drug
Our experimental treatment relies on a soluble form of 17β-estradiol, enabling administration of the drug i.v. or intraosseously. We have examined microencapsulation with β-cyclodextrin, as well as mono- and disulfate-conjugated E2. While all were efficacious, none were available in a good manufacturing practice (GMP) form, as required by the FDA. We were also advised to explore the synthetic estrogen 17-α ethinyl estradiol (EE) with a sulfate at the 3 carbon position (EE-3-SO 4 ). EE has been in use for decades, being originally synthesized in 1938, and subsequently used to produce more potent contraceptives ( ), as well as being applied to hormone replacement therapy ( ), with an excellent safety record for both applications. Thus by using this synthetic form of estradiol, we could increase the drug’s efficacy. We have been granted a patent for the EE-3-SO 4 drug and are pursuing an FDA application for an investigational new drug (IND). Preclinical experimentation is under way for both TBI and severe hemorrhage in our laboratory and with collaborators.
Estrogen and Traumatic Brain Injury
The Model System and Traumatic Brain Injury Induction Protocol
We have elected to use a standard lateral fluid percussion (LFP) model ( ) to induce moderate TBI, which has been described in detail previously ( ). LFP allows for comparisons between the injured ipsilateral site and corresponding uninjured contralateral site. The animal subjects are adult male Sprague–Dawley rats, which weigh from 330 to 370 g. All procedures conform to the guidelines for animal use and care of the DoD or NIH and are reviewed and approved by our University of Alabama at Birmingham Animal Care and Use Committee, as well as the DoD Animal Care and Use Review Office. Surgical procedures are performed aseptically and under isoflurane anesthesia. The preparation of the rats for the TBI proceeds in two stages: first an access point through the skull is created surgically and 24 h later the rats are subjected to the LFP injury. Sham control rats receive the initial craniotomy only.
Lateral Fluid Percussion Induction
LFP uses a hydraulic pressure pulse, introduced into the skull via a surgically fashioned trephine. This creates a direct linkage from the LFP apparatus (VCU Biomedical Engineering, Richmond, VA) to the brain’s exposed dura. The trephine craniotomy site is precisely located stereotaxically to be over the right parietal cortex, midway between bregma and lambda, and tangential to the sagittal suture. The LFP apparatus generates its pulse via a precision pendulum-mounted hammer that strikes a piston, resting in a cylinder filled with ultrapure water, fitted with tubing that delivers the pressure pulse to the rat brain. To have a sealed connection to the brain devoid of leakage, a luer fitting is attached to the skull over the craniotomy site. This female luer fitting is fastened and sealed with dental acrylic and a screw. Also included with the LFP apparatus is an inline pressure transducer which logs the force into a workstation’s spreadsheet. Additional quality control measures monitored were apnea time and the righting reflex, which also correlate with LFP force. Any of these quality control values falling outside the expected range (ie, LFP force, apnea time, and righting reflex) were cause to exclude the rat.
After 45 min has elapsed post-TBI, a cannula is implanted surgically in the femoral vein. At 1 h post-TBI, EE-3-SO 4 drug or vehicle (sterile saline) are administered in a minimal volume (∼50 μl). Our chosen dose is 1 mg/kg body weight, based on the optimum dose for T-H treatment, but there are preliminary hints that 5 mg/kg may be better in treating TBI.
We have investigated other drugs in combination with EE-3-SO 4 [eg, memantine, progesterone, G-1 (a synthetic G-protein estrogen receptor agonist), or glucosamine], but combinations showed nil to modest additivity and were not statistically significant, so combinatorial studies were abandoned (data not shown).
Evaluation of Efficacy for EE-3-SO 4 in Treating Traumatic Brain Injury
We have selected a panel of tests that correlate with clinically important parameters, as well as sensitive measurements which can reveal alleviation of early damage from TBI. Please note that these results will be derived from both E2-SO 4 and EE-3-SO 4 , since EE-3-SO 4 is being tested in the same manner as E2-SO 4. The data to be presented are
Noninvasive, whole organism
Behavioral testing (cognition, memory)
MRI (T2-weighted for edema, and fractional anisotropy for axonal diffusion measurements, ie, axon integrity)
Weight gain and/or loss
Physical activity measurement for 12 h nighttime intervals
Neuronal stretch assay (measures viability of explanted brain cells in vitro after mechanical stress)
Invasive brain physiological measurements
Intracranial pressure (ICP), partial brain oxygen tension (pbtO 2 ), brain perfusion pressure, temperature
Data are presented as mean ± standard error of the mean (SEM), unless otherwise specified. Statistical analysis was conducted with SigmaStat Statistical Advisory Software (v. 3.5, Systat Software Inc). Student’s t -test was used to compare groups. Between groups comparisons were conducting using a one-way analysis of variance (ANOVA) followed by either a Holm-Sidak or Newman–Keuls post hoc analysis, as appropriate. Statistical significance was set at an alpha level p ≤ .05.