Upconversion Nanoparticles and Brain Photobiomodulation: What Is Established and What Is Still Hypothesis
Introduction
Every transcranial photobiomodulation (PBM) protocol runs into the same optical problem. Near-infrared light gets through tissue better than visible light, but it still weakens quickly with depth. Reaching cortical targets is plausible. Deep structures such as the brainstem, thalamus or hippocampus are much harder to reach with a light source outside the head. Researchers have proposed several engineering answers: implanted emitters, optical fibres, and nanomaterials that convert light locally. This article looks at one of them, lanthanide-doped upconversion nanoparticles (UCNPs). It separates what has been shown, mostly in optogenetics, from what has only been proposed for PBM.
This angle is narrow on purpose. The literature is easy to over-read because the optogenetics results are striking, and they are not PBM results. A practitioner who sees “near-infrared deep brain stimulation” in a headline should know what kind of evidence stands behind it.
What upconversion nanoparticles are
Upconversion is an anti-Stokes process. Lanthanide ions held in a crystal host absorb two or more low-energy photons in sequence and emit one photon of higher energy. A review by Meynaghizadeh-Zargar and colleagues in Photobiomodulation, Photomedicine, and Laser Surgery (review of UCNPs for brain PBM) explains why this matters for PBM. Low-energy near-infrared photons get through tissue well but are biologically less potent. Upconversion turns them, at the particle, into higher-energy photons that act more strongly but could not have travelled that far through tissue on their own. The authors discuss brain PBM and the optical windows of brain tissue. They propose that placing UCNPs in deep regions and irradiating them with well-penetrating wavelengths could make brain PBM more efficient.
That paper is a narrative review and a proposal. For PBM it is hypothesis-level evidence, not a clinical finding. The review asserts that the particles cross the blood–brain barrier and have low toxicity, but presents no primary data for either claim. An independent in vitro study (Acta Naturae 2020, in vitro study of UCNPs on brain cells) found that ytterbium- and thulium-doped UCNPs were non-toxic to glioma cells but had a moderate cytotoxic effect on primary neuronal cultures at high concentrations; its abstract does not report blood–brain barrier testing. Brain delivery and safety therefore remain unestablished for any UCNP formulation.
What has been demonstrated: optogenetics in mice
The strongest data come from a neighbouring field. In a 2018 Science paper (2018 Science paper on UCNP optogenetics), Chen and colleagues used UCNPs designed to absorb tissue-penetrating near-infrared light and emit visible light, so that the particles act as optogenetic actuators. In mice, this triggered dopamine release in the ventral tegmental area, induced brain oscillations through the medial septum, silenced seizures, and triggered memory recall. A methods chapter describes the protocol for stimulating dopamine neurons in anesthetized mice (protocol for deep brain stimulation in mice).
These results come from animals, and they are optogenetic. Optogenetics depends on neurons that have been engineered to express a light-gated channel or pump, so the emitted light hits a defined molecular target. PBM has no engineered target of this kind. Its proposed targets, such as mitochondrial cytochrome c oxidase, occur naturally in the body, and which photoacceptor matters in a given setting is still debated. So the mouse work shows that a UCNP can deliver useful visible light to a deep site through intact tissue. It does not show that the same approach produces a PBM effect on the body’s own light-absorbing molecules in a human.
A newer step, still in animals, is a 2025 Science Advances paper (2025 Science Advances study on optoelectronic nanoparticles). The authors used hybrid upconversion-photovoltaic nanoparticles to turn near-infrared light into localized electrical stimulation of neurons, in ordinary mice with no genetic modification. They report modulation of neurons in the medial septum and ventral tegmental area seven days after injection, suppressed seizures, and triggered dopamine release. This is no longer light-based biomodulation in the PBM sense, because the particles act as electrical transducers. It is included here because it shows how fast “nanoparticle plus near-infrared light” is branching into mechanisms that are not PBM at all. The abstract reviewed for this article did not report safety findings, so nothing can be said about them.
A cautionary example of how to read the nanoparticle literature
A 2025 Scientific Reports paper (2025 study of UCNPs in bladder cancer cells) tested red-emitting UCNPs (a NaYbF4 host doped with samarium and neodymium) on T24 bladder cancer cells under 980 nm excitation. According to the abstract, cell toxicity rose with particle concentration, and more so when combined with radiation. The particles raised reactive oxygen species and changed the expression of several genes. The authors read this as a possible cancer therapy.
This study is not PBM evidence. It is included for what it shows about the field. A UCNP that emits red light is not automatically a biostimulatory device. The particle itself can be toxic to cells, cause oxidative stress, and change gene expression, regardless of the light it emits. The same material one group hopes to use against cancer would be an unwelcome result for a group hoping for gentle photobiomodulation. Any claim of UCNP-mediated PBM therefore needs controls that separate the effect of the particle from the effect of its emitted light. The abstract did not report doses, so it is unclear how this study separated those variables.
Unresolved translation problems
A 2026 review in Materials Today Bio (2026 review of UCNPs in biomedicine) names four unsolved obstacles for UCNPs in biomedicine generally: 1. making the particles bright enough while keeping the excitation energy safe for tissue; 2. making particle behaviour in living systems predictable; 3. making sure the particles are cleared safely from the body; 4. improving reproducibility between studies.
These are the authors’ assessment. None of them is unique to PBM, but each matters more for PBM.
Take the excitation problem. PBM works within a fairly narrow dose range, and the field has long argued about biphasic dose responses and about whether the dose that reaches the target is known at all. UCNPs usually convert light inefficiently. A source strong enough to excite them at depth may therefore deposit a lot of energy in the tissue the light passes through on the way. No verified study was found that measures this trade-off for PBM, so it is presented here as a reasoned concern (opinion), not a measured result. Evidence on UCNP dosimetry specific to PBM is currently thin.
Predictable behaviour and clearance also matter clinically. Practitioners accept PBM largely because it is non-invasive and puts no foreign material into the body. A nanoparticle approach changes the risk category: an injected material that may stay in tissue has to be evaluated like a drug or an implanted device. We found no approved UCNP product for PBM anywhere. An injected nanoparticle would not be regulated like a light device alone: in the US, FDA guidance on drug products containing nanomaterials (2022) would likely apply, and in the EU, the Medical Device Regulation (Annex VIII, Rule 19) classifies devices incorporating nanomaterials by their potential for internal exposure: Class III for high or medium potential, Class IIb for low and Class IIa for negligible. The exact pathway would depend on how the product is classified.
What this means for practitioners
First, nothing in this literature supports clinical use of UCNPs for PBM today. Searches of PubMed and Europe PMC found no human trial of UCNP-mediated PBM. The evidence consists of a narrative review proposing the idea, animal experiments in optogenetics and neuromodulation, and cell-line work.
Second, keep the categories clear when you read or talk about this field. Established evidence for PBM comes from conventional light delivered from outside the body. Emerging evidence for deep delivery comes from mouse optogenetics, which works through a different target mechanism. The idea that UCNPs could carry PBM to the brainstem is a hypothesis.
Third, apply the same scepticism about dose that you would to any PBM claim. Ask: 1. what wavelength was delivered; 2. at what irradiance and for how long; 3. how much light actually reached the target; 4. whether the study included particle-only and light-only controls.
Fourth, patients may ask about “nanoparticle light therapy” they have seen online. An honest answer is that it is a laboratory research area, not a clinical option, and that its safety in humans has not been established.
Summary
UCNPs offer a physically sound way to deliver light to tissue that outside sources cannot easily reach. In mice, they have driven deep neuronal activity through the intact skull, but only in optogenetic or electrical-transducer designs. For true PBM, the verifiable literature is limited to a proposal and indirect work. Four questions remain open: excitation dose, particle toxicity, clearance and reproducibility. They need answers from human-relevant data before the idea can inform practice. For now, UCNP-based PBM is a research direction to follow, not a tool to use.
📧 in**@*****************re.com
☎️ +351 915 698 420
OLA Laser Academy
Instructor: Oren Ziv Ram
NIF: 314055314
Online Education in Laser Acupuncture & Auricular Medicine
© 2025 OLA Laser Academy. All rights reserved.