The Pulse and the Protein

There is a scene that repeats itself in molecular biology laboratories around the world, from Boston to São Paulo.

A PhD student walks into the laboratory at seven in the morning, runs a gel, and finds that the band that should be there is missing. She changes the bacterial strain. It does not appear. She lowers the induction temperature, reduces the inducer, adds different protease inhibitors. It still does not appear. Four months later, the protein remains insoluble, like a white paste at the bottom of the tube, and the question that motivated her entire thesis remains exactly where it was.

In laboratories at USP, UFRN, and UNICAMP, this scene repeats itself with different names and the same underlying structure. It is the invisible cost of experimental science: much of the time spent at the bench is dedicated to results that do not come. A significant number of PhD projects reach their conclusion without answering the question that motivated them in the first place.

What is rarely told is what comes next: the answer that emerges instead is often more interesting than the one originally being sought. The literature on electrical stimulation provides one of the clearest examples of this phenomenon in contemporary neuroscience.

Twenty Nanometers

Between one neuron and another lies a gap of approximately 12 to 50 nanometers: the synaptic cleft. It is not empty space. It is one of the most protein-dense environments in the body, containing adhesion molecules that connect both sides of the synapse, including neurexins, neuroligins, and LRRTMs; extracellular matrix components that shape the surrounding environment, such as chondroitin sulfate proteoglycans and perineuronal nets; proteases that remodel this matrix when space needs to be cleared, including MMP-9 and tPA; and diffusible trophic factors, most notably BDNF.

This molecular network helps determine which neural connections survive and adapt during rehabilitation. It also responds to electrical stimulation in ways that have already been experimentally characterized.

What the Electrical Pulse Does to Proteins

1. 1. Peripheral nerve: one hour can produce effects measured in weeks. In 2000, Tessa Gordon and Thomas Brushart’s group applied electrical stimulation at 20 Hz for one hour to the proximal stump of the transected femoral nerve in rats. The effect was measured at the molecular level: BDNF messenger RNA increased threefold, and TrkB receptor RNA doubled within the first eight hours, reaching peaks of sixfold and fourfold, respectively, after two days. Blocking action potentials with tetrodotoxin abolished the effect, demonstrating that the signal was transmitted to the cell body. GAP-43 and tubulin expression also increased, reflecting activation of gene programs associated with regeneration.

1. The surprise: more stimulation was worse. The same group compared one hour of stimulation with three hours, one day, seven days, and fourteen days. In sensory neurons of the dorsal root ganglion, one hour outperformed both the unstimulated control and all longer stimulation periods, contradicting the investigators’ initial expectation. A hypothesis that emerged from subsequent work is that excessive BDNF exposure may desensitize the TrkB receptor, disrupting the pathway that should otherwise be activated. The relationship between stimulation dose and biological response may therefore be non-linear.

The experiment was designed to find the optimal duration of stimulation. It found the opposite of what was expected — and that finding ultimately became part of the path toward clinical therapy.

1. 1. Cortex: direct current requires BDNF. In 2010, Fritsch and colleagues published a study in Neuron that helped clarify the biological mechanisms underlying transcranial direct current stimulation (tDCS). Direct current stimulation of mouse motor cortex slices induced long-term potentiation (LTP) in a polarity-specific and NMDA-receptor-dependent manner, requiring coupling with low-frequency synaptic activity. The decisive finding was that potentiation did not occur in BDNF- or TrkB-mutant mice. Incubating the tissue with TrkB-IgG, a recombinant protein capable of sequestering BDNF, also abolished the effect.

1. 1. From electrophysiology to chromatin. Six years later, twenty minutes of anodal tDCS in mice increased hippocampal LTP, learning, and memory, with effects lasting for approximately one week. The mechanism involved epigenetic regulation: increased acetylation of promoter I of the Bdnf gene, increased binding of pCREB to the promoter, recruitment of CBP, and greater BDNF protein expression. Inhibiting acetylation or blocking TrkB disrupted the molecular, electrophysiological, and behavioral effects.

1. 1. Structure follows function. Using two-photon microscopy in living animals, Gellner and colleagues showed that direct current stimulation combined with peripheral electrical paw stimulation rapidly increased dendritic spine density in the sensorimotor cortex. The effect persisted for 24 hours and was associated with greater survival of existing spines and the formation of new spines with larger heads. In mice expressing approximately half the normal level of BDNF, this structural effect disappeared.

1. 1. The intracellular chain. In spinal cord neurons, electrical stimulation increases BDNF expression through calcium- and Erk-dependent signaling. Nifedipine, which blocks voltage-gated calcium channels, significantly reduced the increase, while an Erk inhibitor abolished it completely.

1. 1. The matrix brake. Not every biological mechanism accelerates plasticity. In rats with complete spinal cord transection treated with epidural stimulation and locomotor training, chondroitin sulfate proteoglycans (CSPGs) progressively increased after injury, and training did not reverse this response. γ-motor neurons, which lack perineuronal nets, underwent substantially greater remodeling than α-motor neurons, which are surrounded by them. Only in α-motor neurons was CSPG expression negatively correlated with putative synaptic contacts.

1. In the clinic. Paired vagus nerve stimulation (VNS), in which electrical pulses are delivered at the precise moment of movement during physical therapy, more than doubled rehabilitation gains in rats and was associated with increased corticospinal connectivity two months after treatment ended. In a pivotal human clinical trial, 47% of treated patients achieved a clinically meaningful response compared with 24% in the sham group. The FDA approved the system in 2021.

Note the pattern. Seven studies, three biological scales, and one recurring element: BDNF and TrkB. It is the protein that the electrical current encounters first.

Manufacturing the Molecule to Be Able to Ask

None of these findings would have been possible without recombinant proteins. It is a technical detail that rarely makes headlines but fundamentally determines what can be measured— and therefore what can be understood.

The TrkB-IgG used to abolish potentiation in the Neuron study is a genetically engineered receptor that functions as a molecular sponge. Without it, the experiment would have demonstrated only correlation: BDNF increases, and plasticity increases. With it, the researchers could demonstrate causality: remove the protein, and the effect of electrical stimulation disappears.

A realistic panel of recombinant neuronal proteins relevant to neuromodulation research includes:

1. 1. Soluble receptors (TrkB-Fc, TrkB-IgG): Used to test whether specific signaling pathways are causally necessary for stimulation-induced effects.
   2. Mature BDNF and proBDNF: Essential standards for ELISA assays capable of distinguishing the two molecular forms in serum, a distinction that can substantially affect clinical interpretation.

1. 1. Neurexin-1β and neuroligin-1 ectodomains: The soluble neurexin ectodomain, released into the synaptic cleft by activity-dependent metalloprotease activity, can trigger postsynaptic calcium influx through N-type channels in immature neurons and NMDA receptors in mature neurons. This provides a direct mechanistic hypothesis linking electrical stimulation to synaptic adhesion remodeling.

1. 1. Chondroitinase ABC: A recombinant enzyme capable of digesting CSPGs and therefore a potential experimental tool for testing whether removing extracellular matrix constraints can amplify the effects of epidural stimulation.

The choice of expression system determines whether an experiment is interpretable, and this is where months of bench work can either be lost or saved. E. coli is fast and inexpensive, but it does not perform mammalian glycosylation and is often poor at forming complex disulfide bonds. Yeast produces different glycosylation patterns. Mammalian systems such as HEK293 and CHO cells are more expensive and generally yield less protein, but they can produce proteins with folding and post-translational modifications more closely resembling those recognized by the human nervous system. For many proteins associated with the synaptic cleft, mammalian expression is therefore the preferred approach.

The Experiment That Has Not Yet Been Done

Technology already exists to directly map the protein composition of the synaptic cleft. Proximity labeling uses enzymes fused to molecular targets positioned within the synaptic environment to biotinylate proteins within an approximately 20-nanometer radius. Foundational work published in Cell in 2016 generated distinct molecular profiles for excitatory and inhibitory synaptic compartments.

What remains to be done is to connect three pieces of the puzzle: apply a documented electrical stimulation protocol, map the resulting proteomic signature of the synaptic environment, and correlate those molecular changes with individual motor recovery rather than relying exclusively on group averages.

In a cohort of 84 patients undergoing subacute stroke rehabilitation, individual increases in mature BDNF were associated with greater reductions in stroke severity, even though the group average did not increase. The signal appeared in individual variation rather than in the aggregate mean.

This creates an important opportunity for neurotechnology research. A rigorous experimental platform can precisely document the parameters of stimulation while molecular biology tools characterize the biological response. Aligning these two layers could help answer a question that remains difficult to resolve: how does a specific electrical stimulus translate into a specific biological and functional response?

The laboratory knows how to measure the molecule. Neurotechnology knows how to measure the pulse. The opportunity is to place the two datasets side by side.

The Wrong Door

Go back to the PhD student with the empty gel.

The history of electrical stimulation suggests that she should pay attention to what is going wrong. Gordon and Brushart were looking for the optimal duration of stimulation and found that less could be more — a finding that only became intelligible after considering the possibility of TrkB receptor desensitization. Fritsch and colleagues sought to explain why tDCS works and ultimately identified a form of synaptic plasticity that depends on a specific protein.

Neither research program was defined by the result its investigators initially expected to find. Their most important contribution emerged from following the unexpected result far enough to understand its mechanism.

The 47% clinically relevant response rate observed in the pivotal vagus nerve stimulation trial also illustrates the importance of this perspective. More than half of the treated patients did not achieve a clinically meaningful response. Those outcomes should not simply be classified as technological failure. They represent biological variation that may contain information about who responds, why they respond, and which mechanisms determine the effectiveness of stimulation.

Understanding those differences may require looking beyond the electrical pulse itself and into the molecular environment that receives it.

Stimulating the nerve is the part we already know how to do. Understanding what happens in the next 20 nanometers may be the work of the next decade.

References

[1] Al-Majed AA, Brushart TM, Gordon T. Electrical stimulation accelerates and increases expression of BDNF and trkB mRNA in regenerating rat femoral motoneurons. European Journal of Neuroscience, 2000;12(12):4381–4390.

[2] Al-Majed AA, Neumann CM, Brushart TM, Gordon T. Brief electrical stimulation promotes the speed and accuracy of motor axonal regeneration. Journal of Neuroscience, 2000;20(7):2602–2608.

[3] Geremia NM, Gordon T, Brushart TM, Al-Majed AA, Verge VMK. Electrical stimulation promotes sensory neuron regeneration and growth-associated gene expression. Experimental Neurology, 2007;205(2):347–359. · Hipótese de dessensibilização do TrkB revista em: Zuo KJ, Gordon T, Chan KM, Borschel GH. Electrical stimulation to enhance peripheral nerve regeneration. Experimental Neurology, 2020;332:113397.

[4] Fritsch B, Reis J, Martinowich K, Schambra HM, Ji Y, Cohen LG, Lu B. Direct current stimulation promotes BDNF-dependent synaptic plasticity: potential implications for motor learning. Neuron, 2010;66(2):198–204. doi:10.1016/j.neuron.2010.03.035

[5] Podda MV, Cocco S, Mastrodonato A, et al. Anodal transcranial direct current stimulation boosts synaptic plasticity and memory in mice via epigenetic regulation of Bdnf expression. Scientific Reports, 2016;6:22180. doi:10.1038/srep22180

[6] Gellner AK, Reis J, Holtick C, Schubert C, Fritsch B. Direct current stimulation-induced synaptic plasticity in the sensorimotor cortex: structure follows function. Brain Stimulation, 2020;13(1):80–88. doi:10.1016/j.brs.2019.07.026

[7] Wenjin W, Wenchao L, Hao Z, et al. Electrical stimulation promotes BDNF expression in spinal cord neurons through Ca²⁺- and Erk-dependent signaling pathways. Cellular and Molecular Neurobiology, 2011;31(3):459–467. doi:10.1007/s10571-010-9639-0

[8] Al’joboori YD, Edgerton VR, Ichiyama RM. Effects of rehabilitation on perineural nets and synaptic plasticity following spinal cord transection. Brain Sciences, 2020;10(11):824. doi:10.3390/brainsci10110824

[9] Meyers EC, Solorzano BR, James J, et al. Vagus nerve stimulation enhances stable plasticity and generalization of stroke recovery. Stroke, 2018;49(3):710–717. doi:10.1161/STROKEAHA.117.019202

[10] Dawson J, Liu CY, Francisco GE, et al. Vagus nerve stimulation paired with rehabilitation for upper limb motor function after ischaemic stroke (VNS-REHAB): a randomised, blinded, pivotal, device trial. The Lancet, 2021;397(10284):1545–1553. doi:10.1016/S0140-6736(21)00475-X · Desfechos de longo prazo: Kimberley TJ, Cramer SC, Wolf SL, et al. Stroke, 2025;56(8):2255–2265.

[11] Wierda KDB, Toft-Bertelsen TL, Gøtzsche CR, et al. The soluble neurexin-1β ectodomain causes calcium influx and augments dendritic outgrowth and synaptic transmission. Scientific Reports, 2020;10(1):18041. doi:10.1038/s41598-020-75047-z

[12] Loh KH, Stawski PS, Draycott AS, et al. Proteomic analysis of unbounded cellular compartments: synaptic clefts. Cell, 2016;166(5):1295–1307. doi:10.1016/j.cell.2016.07.041

[13] Serum BDNF as an objective indicator of rehabilitation response in patients with subacute stroke. Journal of Clinical Medicine, 2026;15(14):5578. doi:10.3390/jcm15145578

[14] Profiles and interviews about Orby.co and the Ortech device: Brazil Journal (2026); Fast Company Brasil (2025); Google Impact.