- BrainSightAI Team
The Future of Neuroimaging- Molecular fMRI
Research by: Ranganayaki Sathyanrayanan, Haritha George and Laina Emmanuel
Edited by: Tanvi Sheth and Simran Rana
Neuroimaging is a powerful tool for both neurologists and psychiatrists to understand neurobiological processes, measure neurotransmitter levels, and characterize disorders and diseases (Ceccarini et al, 2020). With the help of such techniques, it is possible to establish a comprehensive map of brain function that can be used in understanding the complex pathological causes behind psychiatric and neurological disorders (Bartelle, 2016; Liu et al, 2017).
Let’s get an understanding of different imaging techniques and how they developed.
BOLD (Blood Oxygenation Level Dependent) fMRI
BOLD functional Magnetic Resonance Imaging (fMRI) is based on the magnetic properties of the haemoglobin molecule when it binds to oxygen. BOLD fMRI has great advantage in whole-brain neuroimaging but provides very little specificity for neural pathways or signaling components of interest.
However, a plethora of different cells and cell types contribute simultaneously to the observed fMRI data from each 3D volume element and as a result, the cellular origins of functional imaging signals get obscured. Thus, the balance of excitation, inhibition, or neuromodulation that gives rise to localized BOLD responses can not be determined.
fMRI techniques cannot detect blood flow changes accurately as the signal changes they detect are minimal, rendering a low temporal resolution. Consequently, statistical analysis of repeated trials or stimulus presentations makes the technique prone to false positives and negatives.
Researchers have tried to address the limitations of bold-fMRI in the following ways:
Introduced empirical characterization of the relationships between BOLD responses and cellular-level signals detectable by electrophysiology or optical imaging.
Started mechanistic dissection of signaling pathways that relate neural activity to hemodynamic changes.
Mapped fMRI responses to specific brain stimuli of molecular and cellular origin.
In addition to the above, a few radical fMRI approaches have tried to resolve endogenous activity-dependent brain signals arising from non-hemodynamic sources, such as diffusion changes (Le Bihan, 2007), neuronal magnetic fields (Bandettini et al., 2005), and metabolite-dependent spectroscopic signals (Mangia et al., 2009; Hyder and Rothman, 2012).
Functional imaging techniques with improved specificity have been sought by sensitizing MRI acquisition schemes primarily to small blood vessels, trying to tease apart the separate contributions of blood flow, volume, and oxygenation changes to hemodynamic responses using methods such as calibrated fMRI.
BOLD fMRI is based on the magnetic properties of the haemoglobin molecule when it binds to oxygen. (Adapted from Uludağ K., Uğurbil K. (2015) Physiology and Physics of the fMRI Signal.)
Cellular level methods
Cellular-level methods include modern optical imaging like fluorescent imaging, brightfield microscopy, photon microscopy and confocal microscopy. Cellular-level methods, although precise, only address a small portion of mammalian brains (Bartelle et al, 2016).
We now know that connectivity between populations of neurons matters; but whole-brain neuroimaging techniques provide little specificity for neural pathways or for signaling components of interest.
Hence, doctors need techniques that have the potential to combine the specificity of cellular-level measurements with the noninvasive whole-brain coverage of fMRI. That’s where novel techniques like molecular fMRI and PET/MRI come into the picture.
PET/MR is a recently developed method that allows a more thorough examination of brain connectivity and underlying physiological processes (Aiello, Cavaliere & Salvatore, 2016). The combination of these two modalities into a single device merges functional and morphologic information from MR imaging with molecular PET data (Wehrl, 2013).
PET/MR combines the strengths of both the PET and MR imaging techniques. The PET has high detection sensitivity and accurate quantification, but lacks good spatial resolution and tissue contrast. On the other hand, MR imaging enables high-resolution imaging of morphology with good soft-tissue contrast, detects endogenous metabolite distributions using spectroscopy and allows dynamic acquisition of tissue perfusion (Wehrl, 2013).
According to Bartelle et al’s (2016) paper, molecular fMRI is “a hybrid of molecular imaging with fMRI in which targeted molecular probe-mediated readouts form the basis for functional brain imaging”. It uses contrast agents other than haemoglobin that are specific to neural signaling. As a result, molecular fMRI can employ magnetic iron-containing protein ferritin as a reporter gene and allow the expression of a target gene to become detectable using MRI.
Mechanisms of contrast agents for molecular fMRI. Each panel represents a different type of contrast agent, including a typical example of each (left) and its mechanism of influencing MRI signal (right). (Adapted from Bartelle et al, 2016 and Jassanoff, 2007.)
The breakthrough of Molecular fMRI
One of the first landmark studies using molecular fMRI to map dopaminergic brain activity was carried out by Lee et al (2014). The probe (as is used in molecular fMRI) was a magnetically active metalloprotein, which acted like haemoglobin, and binded dopamine selectively. Intermittent electrical stimuli to the medial forebrain bundle helped Lee et al. acquire serial images successfully.
The dopamine imaging revealed a map of peak dopamine concentrations evoked by reward-related stimulation. Prima facie the results resembled that of a basic fMRI, but the results included quantitative measures of an individual signalling molecule. Therefore, molecular fMRI in its early stages of development provided a more sophisticated spatial and temporal resolution in comparison to the standard imaging methods based on PET scans (Lee et al, 2014; Bartelle et al, 2016).
Modern neuroscience is now facing the dual challenge of mapping macro and micro level brain networks, as well as integrating the acquired knowledge of molecular and cellular-level neurobiology into organismic-scale models of brain function in physiology, behaviour and cognition. This neuro technique can be an “increasingly successful approach for spatiotemporally-resolved studies of diverse neural phenomena” (Ghosh et al, 2018).
Advancements in this technology could lead to the determination of neurochemical maps across the width of the brain. A number of electrical, chemical and behavioural stimuli can be used to find such significant activations. Bartelle et al (2016) predict, “such methods might provide fundamental insights into the relationships between neurotransmitter release patterns and cognitive phenomena, affective states, and diseases, with resolution perhaps approaching the level of individual cells” in in-vivo animal, and eventually human clinical studies.
Challenges with molecular fMRI
Preliminary results of molecular fMRI show promise but there are a few challenges that need to be addressed before it becomes a reality (Bartelle, 2016)
Need for effective and minimally invasive methods for delivering imaging agents for large brain volumes
Need for probes that permeate cells for intracellular targets
Ability to get contrast agents across blood-brain barrier seamlessly
Need for probes with higher sensitivity
Data analysis challenges with molecular-fMRI
Concentration level changes from person to person
Challenges related to measuring the baseline
Spurious relations when particular regions have not just one, but multiple neurotransmitter sites or receptors
A larger focus on serotonin and dopamine pathways
A possibility of nuclei with multiple receptors nearby which may not be clearly differentiated in imaging.
Immune system triggers like ACH concentrations and cytokines
Molecular fMRI does seem to have valuable benefits in Psychiatry and Pharmacology and BrainSightAI plans to employ its use in VoxelBox. VoxelBox is an indication-agnostic platform to map an individual’s structural and functional connectome and compare it against the healthy connectome. Over and above it, the platform uses AI to classify the resultant connectomic patterns and anomalies. If you are intrigued by these questions and would like to get involved, please write to us at firstname.lastname@example.org
Living Brain | Stories - Dassault Systèmes®. Accessed June 30, 2021. https://www.3ds.com/stories/living-brain/
Aiello, M., Cavaliere, C. & Salvatore, M. (2016). Hybrid PET/MR Imaging and Brain Connectivity. Frontiers in Neuroscience. Vol. 10. doi: https://doi.org/10.3389/fnins.2016.00064
Bartelle, B. B., Barandov, A., & Jasanoff A. (2016) “Molecular fMRI.” Journal of Neuroscience. Vol 36(15) pp 4139–4148. doi:https://doi.org/10.1523/JNEUROSCI.4050-15.2016
Ceccarini J, Liu H, Laere KV, Morris ED, Sander CY. (2020). Methods for Quantifying Neurotransmitter Dynamics in the Living Brain With PET Imaging. Frontiers in Physiology. Vol 11, pp 792. doi:https://doi.org/10.3389/fphys.2020.00792
Ghosh, S., Harvey, P., Simon, J. C., & Jasanoff, A. (2018). Probing the brain with molecular fMRI. Current opinion in neurobiology, 50, 201–210. https://doi.org/10.1016/j.conb.2018.03.009
Jasanoff A. (2007) MRI contrast agents for functional molecular imaging of brain activity. Curr Opin Neurobiol. Vol 17:593–600. doi: 10.1016/j.conb.2007.11.002.
Lee T., Cai LX., Lelyveld V.S., Hai A., Jasanoff A. (2014) Molecular-level functional magnetic resonance imaging of dopaminergic signaling. Science 344:533–535
Liu, J., Li, M., Pan, Y., Lan, W., Zheng, R., Wu, F. & Wang, J. (2017). Complex Brain Network Analysis and Its Applications to Brain Disorders: A Survey. Complexity. Vol 2017, pp 27. doi: https://doi.org/10.1155/2017/8362741
Markicevic, M., Savvateev, I., Grimm, C. et al. Emerging imaging methods to study whole-brain function in rodent models. Transl Psychiatry 11, 457 (2021). https://doi.org/10.1038/s41398-021-01575-5
Wang, F., Yang, J., Pan, F., Ho, R. C., & Huang, J. H. (2020). Editorial: Neurotransmitters and Emotions. Frontiers in psychology, 11, 21. https://doi.org/10.3389/fpsyg.2020.00021
Wark S, Nowina-Krowicki M. But Why? Generating narratives using provenance. arXiv:1812.08597
Wehrl, H. F., Wiehr, S., Divine, M. R., Gatidis, S., Gullberg, G. T., Maier, F. C., et al. (2014). Preclinical and translational PET/MR imaging. J. Nucl. Med. 55(Suppl. 2), 11S–18S. doi: https://doi.org/10.2967/jnumed.113.129221