Differential effects of amplitude-modulated transcranial focused ultrasound on excitatory and inhibitory neurons
This article has been Reviewed by the following groups
Discuss this preprint
Start a discussion What are Sciety discussions?Listed in
- Evaluated articles (eLife)
Abstract
Although stimulation with ultrasound has been shown to modulate brain activity at multiple scales, it remains unclear whether transcranial focused ultrasound stimulation (tFUS) exerts its influence on specific cell types. Here we propose a novel form of tFUS where a continuous waveform is amplitude modulated (AM) at a slow rate (i.e., 40 Hz) targeting the temporal range of electrophysiological activity: AM-tFUS. We stimulated the rat hippocampus while recording multi-unit activity (MUA) followed by classification of spike waveforms into putative excitatory pyramidal cells and inhibitory interneurons. At low acoustic intensity, AM-tFUS selectively reduced firing rates of inhibitory interneurons. On the other hand, higher intensity AM-tFUS increased firing of putative excitatory neurons with no effect on inhibitory firing. Interestingly, firing rate was unchanged during AM-tFUS at intermediate intensity. Consistent with the observed changes in firing rate, power in the theta band (3-10 Hz) of the local field potential (LFP) decreased at low-intensity, was unchanged at intermediate intensity, and increased at higher intensity. Temperature increases at the AM-tFUS target were limited to 0.2°C. Our findings indicate that inhibitory interneurons exhibit greater sensitivity to ultrasound, and that cell-type specific neuromodulation may be achieved by calibrating the intensity of AM-tFUS.
Article activity feed
-
Reviewer #3:
The authors propose a new method of focused ultrasound (FUS) neuromodulation namely amplitude modulated FUS that they propose can differentially affect inhibitory and excitatory cells depending upon the intensity employed. Parameter selection is an issue for this field and the introduction of new methods for efficacious modulation are highly desirable. However, this paper does not explicitly test AM FUS against existing forms of FUS thus lending no evidence to its efficacy. While the differential effects are interesting in themselves, we gain no insight if AM FUS is the critical factor leading to this.
-
Reviewer #2:
Nguyen et al. developed a novel method of transcranial focused ultrasound stimulation and used it to stimulate anesthetized rats while performing extracellular recordings in the hippocampus. They find that the stimulation has different amplitude-dependent effects on putative inhibitory interneurons and excitatory principal cells. This finding is exciting because it suggests that transcranial ultrasound could be used to specifically reduce or increase firing rates in excitatory or inhibitory neurons in a particular part of the brain (resolution in the mm range). In principle, this could also be applied to humans. Simultaneously measured oscillations of the local field potential, particularly (but not exclusively) in the theta band (3-10 Hz) could also be manipulated in a bidirectional manner depending on the stimulation …
Reviewer #2:
Nguyen et al. developed a novel method of transcranial focused ultrasound stimulation and used it to stimulate anesthetized rats while performing extracellular recordings in the hippocampus. They find that the stimulation has different amplitude-dependent effects on putative inhibitory interneurons and excitatory principal cells. This finding is exciting because it suggests that transcranial ultrasound could be used to specifically reduce or increase firing rates in excitatory or inhibitory neurons in a particular part of the brain (resolution in the mm range). In principle, this could also be applied to humans. Simultaneously measured oscillations of the local field potential, particularly (but not exclusively) in the theta band (3-10 Hz) could also be manipulated in a bidirectional manner depending on the stimulation amplitude. Such cortical oscillations have been strongly linked to a wide range of functions including memory, and the potential to manipulate them in an anatomically precise manner is exciting and could even lead to new therapy approaches. Although it is not new that ultrasound can be used to modulate neuronal activity, this paper reaches a new level of precision by demonstrating that bidirectional effects can in principle be limited to one cell class or one frequency band. Thus, it could provide a great alternative to current methods that either provide much less precision (e.g. transcranial magnetic stimulation) or rely on more invasive methods (e.g. deep brain stimulation) or genetics (e.g. optogenetics).
The study is well-designed with stimulation at 3 different amplitudes applied in the same rat, whereby each 3-minute stimulation is compared to a 3-minute sham session where the transducer is 1 cm above the skull. Baseline sessions before each stimulation and sham session did not show any differences, showing no spillover-effects from the previous stimulus. Effects on brain temperature were also measured and shown to be negligible compared to normal variability.
Low intensity stimuli lead to a reduction in firing rates in putative interneurons and a reduction in theta oscillation power, whereas high intensity stimuli lead to an increase in firing rates in putative principal cells, with intermediate intensities having largely no effect.
In principle, these findings could provide novel insights into the mechanisms underlying ultrasound stimulation, but neither of the two discussed main modes of action (mechanical and thermal) appears consistent with the results. Thus, no model could be offered that might give some insight into the underlying mechanisms of ultrasound modulation of neuronal activity. This might be an issue for future work, and if the results were more robust perhaps this would not matter as much. However, the overall size of the effects appears to be too small to be of practical use as a reliable tool for manipulation of neural circuits. Although the authors show statistical significance, some details of the analysis are not fully clear and may need to be further corrected for multiple testing. It remains unclear if perhaps larger or different effects would be achieved when recording through the skin, without anesthesia, in different brain areas, in differently defined subclasses of neurons or with a different stimulation protocol (frequency, duration, amplitude). Thus, although the technique appears promising, more work is needed.
-
Reviewer #1:
Major points:
On the conceptual level, the authors claim that low-intensity amplitude-modulated transcranial focused ultrasound stimulation (AM-tFUS) inhibits local inhibitory interneurons and excites excitatory neurons at high intensity. However, the problem I have with this is that these cell types are highly interconnected within the local circuits, and changing the activity of the inhibitory cell type should have the opposite effect on the excitatory cell type. This has been documented in many experiments (Babl et al., Cell Reports, 2019; Royer et al., Europ.Journ. of Neurosc., 2010), and it unclear why the authors did not see similar effects. Furthermore, it is particularly troubling that the authors observe sustained suppression (five minutes) of the inhibitory neurons yet fail to see any effects on the excitatory …
Reviewer #1:
Major points:
On the conceptual level, the authors claim that low-intensity amplitude-modulated transcranial focused ultrasound stimulation (AM-tFUS) inhibits local inhibitory interneurons and excites excitatory neurons at high intensity. However, the problem I have with this is that these cell types are highly interconnected within the local circuits, and changing the activity of the inhibitory cell type should have the opposite effect on the excitatory cell type. This has been documented in many experiments (Babl et al., Cell Reports, 2019; Royer et al., Europ.Journ. of Neurosc., 2010), and it unclear why the authors did not see similar effects. Furthermore, it is particularly troubling that the authors observe sustained suppression (five minutes) of the inhibitory neurons yet fail to see any effects on the excitatory neurons (Fig. 3B,D). This conceptual problem raises questions about the experimental setup, which I address below.
The authors performed electrophysiological recordings while delivering AM-tFUS with different intensities. To claim the differential effects on the excitatory and inhibitory interneurons, the authors first need to isolate single units in their recordings. However, the authors fail to cluster single units, as documented in the methods section (line 338). There could be several reasons why the authors failed to complete this step. I suggest the following ways to remedy this problem: A. The authors should use a silicone probe with a higher density of recording sites (the distance between the individual sites can be as small as 25 um in some NeuroNexus probes) than the one used in the MS, or use a Neuropixels probe so that the clustering algorithms have a chance to isolate single units. Using NeuroNexus probes with 100 um separation between the recording sites makes it impossible for different channels to "see" the same neuron and severely limits the spike sorting algorithms that separate units based on their unique spatio-temporal waveforms. B. After clustering, the authors should use autoccorellograms to verify that the single units do not violate the refractory period (Hill et al., Journal of Neuroscience, 2011). This is particularly important in areas, such as the hippocampus, which has a high density of neurons, and care should be taken to avoid multiunit recordings. C. The authors should perform one long recording session that comprises all experimental manipulations-the delivery of AM-tFUS, the sham control, and the rest period-to trace how the same units change their firing rate as a function of the experimental manipulations. This would also be very helpful in understanding how the firing rate change in one class of neurons is accompanied by changes in another class. D. Although this might be tricky, the authors could try to perform electrophysiological recordings by lowering the electrode perpendicular to the brain surface. This would allow them to record excitatory neurons and inhibitory interneurons that are connected to each other within the local circuit. This type of recording, would give the authors a greater chance of observing how changes in the firing of the inhibitory cell type affects the activity of the excitatory cell type and vice versa. This type of recording would also be highly desirable for understanding changes in oscillations of the local field potential (LFP) (see below).
The authors should report the sites that they have recorded by labelling the electrode with fluorescent dye or performing lesions at the recording sites.
When analyzing the effect of AM-tFUS on theta frequency oscillations, the authors should perform current source density (CSD) analysis to verify that the observed effects are local and do not originate from distant sources by volume conduction (Buzsaki et al., Nat. Rev. Neurosc. 2016). Performing electrophysiological recordings perpendicular to the brain surface, as I recommend in 2D, would be necessary for this. The CSD analysis would identify the location in the hippocampus where the change in theta power occurs.
The authors argue that temperature changes of 0.2 degrees were not sufficient to alter the firing rate of the neurons. However, the paper to which they refer (Darrow et al., Brain Stim, 2019) shows, in Fig. 7, that heating up brain tissue with a laser even at 0.2C can induce changes in somatosensory evoked LFPs. The authors should perform control experiments that are analogous to those in the cited paper to manipulate the temperature while recording the neurons in order to verify that the observed effects are not due to the changes in temperature.
Minor points:
The authors should not use label cells in Fig. 3 as they cannot claim that they recorded single units.
In Fig. 5C, Fig. S3B,C, and Fig. S4B,C, the authors should show the full scale of the values. Furthermore, the outliers in these plots (not seen in the figures) may drive the general trends, and removing them should be considered.
During AM-tFUS at intermediate power intensity (Fig. 4D,G), the authors observe a very dramatic change in LFP power in the 1-3 Hz frequency range. Although there is no clear underlying change in the firing of neurons at this intensity (Fig. 3E,F,G,H), the authors could speculate on what is happening in this case.
Fig. 5B shows a clear reduction of power in the theta frequency range after AM-tFUS in the dentate gyrus as well as in CA1 and CA3. This effect is also seen in Fig. 4G and Fig. S1,2. Although this effect does not reach the level of statistical significance, the authors should report the p-values.
Although the suppression of firing rates for a five-minute period after low-intensity AM-tFUS application is interesting, I am not sure if such prolonged after-stimulation effects have ever been documented using other modes of neuromodulation. Therefore, the authors should discuss this effect in line with previous work.
-
