Cortical Ripples During NREM Sleep And Waking in Humans Part 3

PY firing precedes IN firing at cortical ripple peaks

Using microelectrode array recordings from granular/supragranular layers of the lateral temporal cortex during NREM, we detected ripples (Fig. 11A) as well as action potentials, which were sorted into those arising from putative single PY or IN units. The mean 6 SD ripple oscillation frequency across microelectrode channels (N = 72) was 89.6 6 1.4 Hz, and across individual ripples (N = 50,967) was 90.1 6 5.2 Hz. We found that INs had a strong tendency to fire at the peak of the ripple, whereas PYs fired shortly before (Fig. 11B). 

In the organizational structure of the human brain, the temporal lobe is considered an important area closely related to memory. The granular layer in the temporal cortex is thought to play a crucial role in this process. Especially in the key short-term memory process, the activity level and number of the granular layer of the temporal lobe cortex will directly determine the individual's memory performance. Therefore, keeping temporal cortical granules healthy is important for our brain and memory.

So, how do you keep temporal cortical granules healthy? First, adequate sleep can promote the growth of brain neurons and enhance the stability of neural circuits, thereby increasing the number and activity of temporal cortex granules. In addition, more exercise, more learning, and more social interaction can also stimulate neuronal activity in the brain and increase the number of neurons, which is beneficial to the health of the temporal lobe cortex granules.

In addition to the above methods, you can also maintain the health of the temporal cortex particles through good eating habits. Studies have shown that some nutrients in foods such as blueberries, walnuts, salmon, and cod can help improve memory and cognitive abilities, and can also slow down the functional decline of temporal cortex granules.

In short, temporal lobe cortical granules are closely related to memory, so it is very important to maintain the health of temporal lobe cortical granules. Exercising more, studying more, socializing more, getting enough sleep, and good eating habits all help to increase the number and activity of granules in the temporal lobe cortex, thereby improving our cognitive ability and memory. It can be seen that we need to improve memory, and Cistanche deserticola can significantly improve memory because Cistanche deserticola is a traditional Chinese medicinal material that has many unique effects, one of which is to improve memory. The efficacy of minced meat comes from the various active ingredients it contains, including acid, polysaccharides, flavonoids, etc. These ingredients can promote brain health in a variety of ways.

Click Know to improve short-term memory

We have previously shown that, in humans, upstates (Csercsa et al., 2010) and spindles (Dickey et al., 2021) are associated with increased unit firing rates. Since cortical ripples are precisely coupled to these events (Fig. 10), their occurrence on upstates and spindles implies an underlying phasic depolarization, which can generate;90 Hz oscillations in computational and experimental models via PY–IN feedback inhibition (Bazhenov et al., 2008; Buzsáki, 2015). Depolarization causes basket cells to fire synchronously, inhibiting other basket cells and pyramids via GABAA. Pyramids fire on recovery, exciting basket cells as they recover, leading to another cycle. 

As this model would predict, we found that PYs and INs were strongly phase-locked to cortical ripples (Fig. 11B), with PY significantly leading to IN spiking (Fig. 11C, D). Furthermore, INs fired at the ripple peak, when pyramidal somatic inhibition would be maximal, as found in cats (Grenier et al., 2001). Similarly, ripple amplitude was higher in waking than NREM (Figs. 3B, 8C), consistent with relative depolarization of pyramidal membrane potential during waking. Phasic depolarization during waking ripples was also suggested by increased .200 Hz amplitude (Figs. 2B, J, 3E, 8F). Thus, human cortical ripples are associated with strong tonic and phase-locked increases in PY and firing and are likely generated by a sudden depolarization triggering PY–IN feedback oscillations (Fig. 11E).

Cortical ripples group cofiring with timing optimal for STDP

The increased tonic and phase-locked unit firing and occurrence of ripples on spindles on the down-to-upstate transition suggest that ripples may facilitate cortical plasticity (Dickey et al., 2021). This could occur through short latency cofiring between units that lead to STDP. Indeed, we found a large increase during ripples of short latency (,5 ms) cofiring between PY–PY unit pairs and PY–IN unit pairs (Fig. 11F). Specifically, there was an increase in cofiring during ripples versus nonripples (PY–PY: 7.16-fold increase, p = 2
10159, N = 4638, t = 28.0; PY–IN: 9.78-fold increase, p = 8
1099, N = 1561, t = 22.7; one-sided paired t-test). Nonripple comparison periods were randomly selected epochs between ripples matched in number and duration.

To test whether this increase in cofiring was not simply because of an increase in the overall firing rates of units during ripples, or even their rhythmicity, but rather a specific organization of cofiring by the phase of individual ripple cycles, we constructed a control dataset consisting of the neuronal firing during ripples where the spike times were randomly shuffled. Compared with this shuffled control, there was a significant increase in short-latency cofiring during ripples with their actual spike times (PY–PY: 1.21-fold increase, p = 1
107, N = 4638, t = 5.2; PY–IN: 1.15-fold increase, p = 1
107, N = 1561, t = 5.2, one-sided paired t-test). This finding indicates that the increase in cofiring is because of a specific organization of cofiring by the phase of the individual ripple cycles. Thus, ripples create a necessary and sufficient condition for STDP and therefore may underlie the crucial contribution of these nested waves to consolidation (Dickey et al., 2021).

Discussion

The current study provides the first comprehensive characterization of cortical ripples during waking and NREM in humans. Using intracranial recordings, we report that human cortical ripples are widespread during NREM in addition to waking. Basic ripple characteristics (occurrence rate, amplitude, oscillation frequency, duration, distribution across cortical areas) are stereotyped across electrodes and cortical parcels. They are also very similar in cortex versus hippocampus, and in NREM versus waking. Cortical ripple stereotypy and ubiquity across structures and states suggest that they may constitute a distinct and functionally important neurophysiological entity.

The characteristics of human ripples are similar to those of rodents with two exceptions. First, the center frequency of human ripples is focally;90 Hz (Jiang et al., 2019a), whereas in rodents it is;130 Hz (Khodagholy et al., 2017). Conceivably, the larger human brain may require more time for corticocortical and hippocampo-cortical integration than the rodent brain. The second apparent difference between humans and rodents is in cortical distribution. A previous rodent study found that cortical ripples were restricted to the association cortex (Khodagholy et al., 2017), and consistent observations were reported in humans (Vaz et al., 2019). In contrast, we recorded cortical ripples in all areas sampled, with slightly lower occurrence rates in association areas as indicated by the positive correlation between occurrence rate and myelination index.

In contrast to the similarities of basic properties, we found that ripples in waking versus NREM occur within many different immediate physiological contexts. In NREM, cortical ripples are strongly associated with local downstates and upstates, and less strongly with sleep spindles. Characteristically, ripples occur on the upslope;100 ms before the peak of the upstate. Previous studies found that sleep spindles (Dickey et al., 2021) and upstates (Csercsa et al., 2010) are associated with strong increases in local unit firing, and we found that also to be the case for cortical ripples during NREM. Hippocampal ripples during NREM in previous studies were also found often to occur during local sharp waves or spindles (Staresina et al., 2015; Jiang et al., 2019a,b,c).

Since sleep spindles and down-to-upstates are characteristic of NREM, they would not be expected to occur with waking ripples. Although we found no other lower-frequency wave to be consistently associated with ripples during waking, both cortical and hippocampal ripples occurred during greatly increased local .200 Hz amplitude activity, a surrogate for unit firing (Mukamel et al., 2005). Thus, the local contexts of both cortical and hippocampal ripples, in both NREM and waking, are sudden increases in local excitability lasting at least as long as the ripple (;70 ms). NREM and waking are different in that the depolarizing pulse is organized by endogenous sleep rhythms in NREM, but appears to be related to exogenous input, such as a retrieval cue, during waking.

Computational neural models and experimental studies have found that depolarizing pulses can induce oscillations from;20 to 160 Hz (Buzsáki and Wang, 2012; Buzsáki, 2015). Depolarization causes synchronous PY firing, which is inhibited for a fixed time by recurrent inhibition from local basket cells. The PYs then fire again, resulting in Pyramidal Interneuron Network Gamma. As predicted by this model, we found strong ripple phase modulation of putative PY and interneuronal firing, with PYs regularly preceding INs. Our findings are also consistent with Pyramidal Interneuron Network Gamma supplemented by synchronized basket cell firing and mutual inhibition (Interneuron Network Gamma) (Bartos et al., 2002). Definitive mechanistic demonstration would require experiments not currently possible in vivo in humans (Stark et al., 2014).

Previous studies have shown that cortical ripples occur on upstate and spindle peaks in cats (Grenier et al., 2001) and rats (Khodagholy et al., 2017), and parahippocampal g bursts of various frequencies occur on upstates in humans (Le Van Quyen et al., 2010). Here we report similar relationships in humans, consistent with a previous finding that cortical ripples during human NREM are suppressed during and increased following downstates (von Ellenrieder et al., 2016). 

The critical role of rodent hippocampal ripples in memory consolidation during NREM (Girardeau et al., 2009) is dependent on their association with cortical sleep waves (spindles, downstates, upstates) (Siapas and Wilson, 1998; Maigret et al., 2016; Latchoumane et al., 2017). In humans, hippocampal ripples are also associated with sleep waves (Staresina et al., 2015; Latchoumane et al., 2017; Jiang et al., 2019a,b), and cortical sleep waves are also associated with consolidation (Niknazar et al., 2015). Thus, our observation that human cortical ripples during NREM are strongly associated with upstates, and less strongly with downstates and spindles, is consistent with the role of human cortical ripples in consolidation. This is supported by a recent study showing that cortical ripples during NREM mark the replay of learned motor patterns from prior waking (Rubin et al., 2022).

Consolidation requires plasticity to increase the strength of the connections embodying the memory, which may occur when presynaptic and postsynaptic cells fire in close temporal proximity, termed STDP (Feldman, 2012). We show here that local neurons are much more likely to fire at delays optimal for STDP during ripples than control periods, and that cofiring is organized beyond what would be expected by a general increase in neuron firing. A similar increase is also observed during sleep spindles and upstates in humans (Dickey et al., 2021). Since we also show that NREM cortical ripples are temporally coordinated with sleep spindles and upstates, this supports synergistic facilitation of plasticity. Thus, multiple characteristics of cortical ripples are consistent with them playing a role in consolidation, but direct confirmation of this role will require interventions that were not performed in the current study.

Recently, human waking cortical ripples were shown to mark spatiotemporal firing patterns during cued recall of items that reproduced those previously evoked by the same items during their initial presentation (Jiang et al., 2017). This suggests the possibility that ripples, in humans and rodents, NREM and waking, hippocampus and cortex, share a common role in contributing to the reconstruction of previously occurring firing patterns. The similar characteristics of human ripples, in NREM and waking, hippocampus, and cortex, are consistent with this speculation.

Previous reports in rodents (Khodagholy et al., 2017) and humans (Vaz et al., 2019) that cortical ripples were restricted to association cortex were also interpreted as consistent with a selective interaction of cortical ripples with the hippocampus. However, in our more extensive sample of cortical sites, we observed a lower ripple density in the higher associative cortex. Furthermore, we found no relationship across cortical parcels between their degree of connectivity with the hippocampal formation (inferred from diffusion imaging) (Rosen and Halgren, 2021) and cortical ripple density or any other ripple characteristic. Indeed, in other work, we have shown that cortical ripples are more likely to co-occur and phase-lock with other cortical sites than the hippocampus (Dickey et al., 2022). Thus, the functional role of cortical ripples may not be confined to memory consolidation and recall.

In conclusion, the current study provides the first report of cortical ripples during sleep in humans. Cortical ripples during NREM were found to have basic characteristics (duration, oscillation frequency, and occurrence rate) that are highly similar to those of cortical ripples during waking as well as hippocampal ripples. Cortical ripple oscillation frequencies were tightly focused at;90 Hz. This finding, together with their stereotypy and ubiquity across structures and states, suggests that cortical ripples may constitute a distinct and functionally important neurophysiological entity. 

Unit firing during human cortical NREM ripples supported the mechanism proposed for rodent hippocampal ripples, based on PY–IN feedback. Cortical ripples during NREM characteristically occur after downstates, during spindles, and shortly before the peak of the upstate, an association and timing that are important for the consolidation of memories during sleep. Furthermore, NREM cortical ripples were associated with increased local cofiring between units, thus fulfilling a fundamental requirement for STDP. However, cortical ripples were widespread across cortical parcels, regardless of their probable connectivity with the hippocampus. Overall, these characteristics of cortical ripples during human NREM are consistent with a role that includes but may not be limited to sleep-dependent memory consolidation.

Reference

1. Axmacher N, Elger CE, Fell J (2008) Ripples in the medial temporal lobe are relevant for human memory consolidation. Brain 131:1806–1817.

2. Bartos M, Vida I, Frotscher M, Meyer A, Monyer H, Geiger JR, Jonas P (2002) Fast synaptic inhibition promotes synchronized gamma oscillations in hippocampal interneuron networks. Proc Natl Acad Sci USA 99:13222–13227.

3. Bazhenov M, Rulkov NF, Timofeev I (2008) Effect of synaptic connectivity on long-range synchronization of fast cortical oscillations. J Neurophysiol 100:1562–1575.

4. Bechtold B (2016) Violin plots for Matlab. Github Project.

5. Behrens TE, Berg HJ, Jbabdi S, Rushworth MF, Woolrich MW (2007) Probabilistic diffusion tractography with multiple fiber orientations: what can we gain? Neuroimage 34:144–155.

6. Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: a practical and powerful approach to multiple testing. J R Stat Soc B 57:289–300.

7. Berens P (2009) CircStat: a MATLAB toolbox for circular statistics. J Stat Softw 31:1–21.

8. Bragin A, Engel J Jr, Wilson CL, Fried I, Buzsáki G (1999) High-frequency oscillations in the human brain. Hippocampus 9:137–142.

9. Buzsáki G (2015) Hippocampal sharp wave-ripple: a cognitive biomarker for episodic memory and planning. Hippocampus 25:1073–1188.

10.Buzsáki G, Wang XJ (2012) Mechanisms of gamma oscillations. Annu Rev Neurosci 35:203–225.

For more information:1950477648nn@gmail.com