Transcranial magnetic stimulation elicits coupled neural and hemodynamic consequences.
ABSTRACT Transcranial magnetic stimulation (TMS) is an increasingly common technique used to selectively modify neural processing. However, application of TMS is limited by uncertainty concerning its physiological effects. We applied TMS to the cat visual cortex and evaluated the neural and hemodynamic consequences. Short TMS pulse trains elicited initial activation (approximately 1 minute) and prolonged suppression (5 to 10 minutes) of neural responses. Furthermore, TMS disrupted the temporal structure of activity by altering phase relationships between neural signals. Despite the complexity of this response, neural changes were faithfully reflected in hemodynamic signals; quantitative coupling was present over a range of stimulation parameters. These results demonstrate long-lasting neural responses to TMS and support the use of hemodynamic-based neuroimaging to effectively monitor these changes over time.
- SourceAvailable from: Stefano Ferraina[Show abstract] [Hide abstract]
ABSTRACT: Recently, neuromodulation techniques based on the use of repetitive transcranial magnetic stimulation (rTMS) have been proposed as a non-invasive and efficient method to induce in vivo long-term potentiation (LTP)-like aftereffects. However, the exact impact of rTMS-induced perturbations on the dynamics of neuronal population activity is not well understood. Here, in two monkeys, we examine changes in the oscillatory activity of the sensorimotor cortex following an intermittent theta burst stimulation (iTBS) protocol. We first probed iTBS modulatory effects by testing the iTBS-induced facilitation of somatosensory evoked potentials (SEP). Then, we examined the frequency information of the electrocorticographic signal, obtained using a custom-made miniaturised multi-electrode array for electrocorticography, after real or sham iTBS. We observed that iTBS induced facilitation of SEPs and influenced spectral components of the signal, in both animals. The latter effect was more prominent on the θ band (4-8 Hz) and the high γ band (55-90 Hz), de-potentiated and potentiated respectively. We additionally found that the multi-electrode array uniformity of β (13-26 Hz) and high γ bands were also afflicted by iTBS. Our study suggests that enhanced cortical excitability promoted by iTBS parallels a dynamic reorganisation of the interested neural network. The effect in the γ band suggests a transient local modulation, possibly at the level of synaptic strength in interneurons. The effect in the θ band suggests the disruption of temporal coordination on larger spatial scales.PLoS ONE 11/2014; 9(11):e112504. · 3.53 Impact Factor
- [Show abstract] [Hide abstract]
ABSTRACT: Cognitive processes require working memory (WM) that involves a brief period of memory retention known as the delay period. Elevated delay-period activity in the medial prefrontal cortex (mPFC) has been observed, but its functional role in WM tasks remains unclear. We optogenetically suppressed or enhanced activity of pyramidal neurons in mouse mPFC during the delay period. Behavioral performance was impaired during the learning phase but not after the mice were well trained. Delay-period mPFC activity appeared to be more important in memory retention than in inhibitory control, decision-making, or motor selection. Furthermore, endogenous delay-period mPFC activity showed more prominent modulation that correlated with memory retention and behavioral performance. Thus, properly regulated mPFC delay-period activity is critical for information retention during learning of a WM task.Science 10/2014; 346(6208):458-63. · 31.48 Impact Factor
- Jung Journal Culture and Psyche 05/2014; 8(2):20-35.
28 September 2007
Vol. 317 No. 5846
Pages 1918 - 1921
Transcranial Magnetic Stimulation Elicits Coupled Neural and
?Elena A. Allen, Brian N. Pasley, Thang Duong, Ralph D. Freeman
Uncovering the Magic in Magnetic Brain Stimulation
? Greg Miller
Supporting Online Material
? Elena A. Allen, Brian N. Pasley, Thang Duong, Ralph D. Freeman
I mage: Depiction of the visual cortex, including
neural, glial and vascular elements, during
transcranial magnetic stimulation (TMS). In this
non-invasive brain stimulation technique, pulses
of current (arrows) are passed through a
figure-eight shaped coil placed above the scalp.
The induced electric field elicits long-lasting
alterations in neural activity which can be
measured with hemodynamic-based imaging
methods. Image: Elena Allen
frequency, because it is selected for as a
recessive in males. Third, all commercially avail-
able products in Europe contain the same CpGV
isolate, which has high genetic homogeneity
(13). Because organic apple growers rely heavily
season, most of the organic orchards are contin-
uously exposed to this virus isolate. Moreover,
each OB of CpGV contains a single virion, in
contrast to nucleopolyhedroviruses with up to
several hundred virions per OB (14). The
potential resistance-delaying effect of a mixture
of virus genotypes in a single infection would
thus be much weaker for CpGV.
The aim of insecticide-resistance manage-
ment is to prevent or delay the selection of
resistance by controlling the factors affecting
allele frequencies in field populations. Our
results make clear that this area of applied
evolutionary biology is also highly relevant to
the application of baculoviruses as biological
control agents. Implementation of resistance
monitoring and resistance management will be
needed in order to sustain the ecological and
economic benefits of this environmentally
friendly class of biological insecticides.
References and Notes
1. D. Mota-Sanchez, P. S. Bills, M. E. Whalon, in Pesticides
in Agriculture and the Environment, W. B. Wheeler,
Ed. (Dekker, New York, 2002), pp. 241–272.
2. J. Mallet, Trends Ecol. Evol. 4, 336 (1989).
3. F. Moscardi, Annu. Rev. Entomol. 44, 257 (1999).
4. D. T. Briese, in The Biology of Baculoviruses, vol. 2,
Practical Application for Insect Control, R. R. Granados,
B. A. Federici, Eds. (CRC Press, Boca Raton, FL, 1986),
5. J. R. Fuxa, in Parasites and Pathogens of Insects, vol. 2,
Pathogens, N. Beckage, S. Thompson, B. Federici, Eds.
(Academic Press, San Diego, CA, 1993), pp. 197–209.
6. F. R. Hunter-Fujita, P. E. Entwistle, H. F. Evans, N. E. Crook,
Eds., Insect Viruses and Pest Management (Wiley, New
7. E. Fritsch, K. Undorf-Spahn, J. Kienzle, C. P. W. Zebitz,
8. B. Sauphanor et al., Phytoma Déf. Vég. 590, 24 (2006).
9. Y. Tanada, J. Insect Pathol. 6, 378 (1964).
10. J. Huber, Mitt. Dtsch. Ges. Allg. Angew. Entomol. 2, 141
11. K. E. Eberle, J. A. Jehle, J. Invertebr. Pathol. 93, 201
12. I. Fuková, P. Nguyen, F. Marec, Genome 48, 1083
13. N. E. Crook, J. D. James, I. R. L. Smith, D. Winstanley,
J. Gen. Virol. 78, 965 (1997).
14. D. A. Theilmann et al., in Virus Taxonomy: The Eighth
Report of the International Committee on Taxonomy of
Viruses, C. M. Fauquet, M. A. Mayo, J. Maniloff,
U. Desselberger, L. A. Ball, Eds. (Elsevier–Academic Press,
New York, 2005), pp. 177–185.
15. We thank B. Wahl-Ermel, A. Wilhelmy, U. Poh, and
E. Gabres for excellent technical assistance in insect
rearing. G. Schöfl, A. Groot, A. Papanicolaou, and
I. Fuková are acknowledged for advice and assistance on
the larval sex-determination methods. This work was
supported by a grant of the Federal Organic Farming
Scheme (05OE023/1) by the Federal Agency for Agriculture
and Food (BLE) of Germany, and by the Max-Planck-Society.
Supporting Online Material
Methods and Materials
Tables S1 and S2
15 June 2007; accepted 20 August 2007
Transcranial Magnetic Stimulation
Elicits Coupled Neural and
Elena A. Allen,* Brian N. Pasley,* Thang Duong, Ralph D. Freeman†
Transcranial magnetic stimulation (TMS) is an increasingly common technique used to selectively
modify neural processing. However, application of TMS is limited by uncertainty concerning its
physiological effects. We applied TMS to the cat visual cortex and evaluated the neural and
hemodynamic consequences. Short TMS pulse trains elicited initial activation (~1 minute) and
prolonged suppression (5 to 10 minutes) of neural responses. Furthermore, TMS disrupted the
temporal structure of activity by altering phase relationships between neural signals. Despite the
complexity of this response, neural changes were faithfully reflected in hemodynamic signals;
quantitative coupling was present over a range of stimulation parameters. These results
demonstrate long-lasting neural responses to TMS and support the use of hemodynamic-based
neuroimaging to effectively monitor these changes over time.
cal, surgical, and pharmacological approaches
(1). In general, these techniques may be invasive,
irreversible, and not confined to specific brain
areas. In contrast, transcranial magnetic stimula-
tion (TMS) (2) provides a noninvasive, revers-
ible, and relatively localized approach that has
substantial promise for basic neuroscience and
clinical applications (3, 4). In this technique, a
magnetic coil placed above the scalp generates
electric currents in the underlying cortex. As yet,
he study of brain function makes use of
various techniques to modify neural pro-
cessing. These include neurophysiologi-
the manner in which these currents affect neuro-
nal processing is largely undetermined (3, 5).
The full potential of TMS depends not only
on a basic understanding of its neural effects,
but also on the ability to make direct measure-
ments of these changes in the human brain. This
has recently been attempted by combining TMS
with noninvasive brain-imaging techniques such
as functional magnetic resonance imaging (fMRI)
and positron emission tomography (PET) (6).
These methods measure hemodynamics and
metabolism to infer changes in neural activity
based on known coupling between these varia-
bles (7). However, in certain conditions, neural
activity may be uncoupled from local hemo-
dynamics. For example, altered brain states such
as seizures (8) and cortical spreading depression
(9) result in complex and atypical physiological
responses that do not fit standard models of
neurovascular coupling. It is essential, therefore,
to investigate both the direct neural effects of
TMS and the relationships among neural, vascu-
lar, and metabolic parameters.
To provide an integrated view of the basic
effects of TMS, we used several complementary
techniques in a controlled physiological prepa-
ration. We applied short TMS pulse trains to the
visual cortex of the anesthetized cat (n = 8) while
simultaneously measuring tissue oxygen and
neural activity (10–12). In separate experiments,
we used 570-nm optical imaging of intrinsic
signals to measure changes in total hemoglobin
(Hbt) within the cortical vasculature (12–14).
Each trial in our experimental paradigm (Fig. 1)
included a pre-TMS baseline (40 s), a short
TMS pulse train (1 to 4 s, 1 to 8 Hz), and a long
recovery period (5 to 15 min). Throughout the
trial, we alternated visual stimulation with a
blank screen to assess the effects of TMS on
both evoked and ongoing spontaneous activity
The neural effects of TMS application are
shown in Fig. 2, A and B. An initial repeated-
measures analysis of variance (ANOVA) on
firing rate indicated significant main effects for
activity state (spontaneous versus evoked,
F241,484= 65.073, P < 10−13) and time (1 to
20 s after TMS, 30 to 90 s, or 180 to 210 s,
F241,484 = 3.473, P < 0.05), as well as a
significant interaction between these factors
(F241,484= 9.931, P < 0.0001) (12). Accordingly,
post hoc tests (Wilcoxon signed-rank) revealed
differential response time courses between ac-
tivity states. Across the population, the sponta-
neous spike rate increased substantially (~200%)
immediately after TMS (Fig. 2A, left) and
remained elevated for ~60 s (P < 0.001; fig.
Helen Wills Neuroscience Institute, Group in Vision
Science, School of Optometry, University of California,
Berkeley, CA 94720, USA.
*These authors contributed equally to this work.
†To whom correspondence should be addressed. E-mail:
28 SEPTEMBER 2007 VOL 317
S4A; fig. S5A, left). In contrast, the evoked
firing rate (Fig. 2A, right) showed an immediate
decrease (~50%) and remained significantly
suppressed for more than 5 min (P < 0.0001;
fig. S4B; fig. S5A, middle). Analogous changes
occurred in the power of local field potentials
(LFPs) (Fig. 2B), although a distinction was
evident with regard to frequency band (fig.
S5B). Spontaneous LFP power at higher
ment, whereas lower frequencies (<40 Hz) ex-
hibited a prolonged reduction, similar to evoked
activity. This distinction is likely related to the
different physiological processes reflected by
these frequency ranges (12, 16).
To determine how neural changes are re-
flected in metabolic and vascular signals, we
examined measurements of tissue oxygen colo-
calized with the neural recordings. A repeated-
measures ANOVA showed a significant main
effect for time (F111,224= 56.609, P < 10−16) but
no effect for activity state (F111,224= 0.0001,
P > 0.98; fig. S6B). Therefore, oxygen was
further analyzed as a single continuous variable
(12, 17). Post hoc Wilcoxon signed-rank tests
revealed a biphasic response pattern for oxygen
(Fig. 2C). An immediate increase peaked at 10
to 15 s after TMS (Fig. 2C, inset; P < 0.001)
and was followed by an extended reduction
lasting over 2 min (P < 0.01). Separate mea-
surements of Hbt (Fig. 2D) revealed a similar
and a subsequent prolonged decrease (over
1 min, P < 0.001). This independent data set
confirms that changes in blood flow underlie a
substantial component of the oxygen response.
main effects of pulse frequency (1,4,or 8 Hz) on
neural (F241,484= 3.522, P < 0.05) and oxygen
(F111,224= 5.739, P < 0.005) data. This raises the
components covary with stimulation parameters.
increase in TMS pulse frequency caused a
monotonic increase in the amplitude of the early
oxygen peak and the level of spontaneous neural
firing (Fig. 3A, upper right quadrant). At later
time points (30 to 90 s), reductions in both tissue
oxygen and evoked spiking were larger with
Fig. 1. TMS and visual
stimulation paradigm. (A)
Timeline of a sample trial
showing stimulus presenta-
tions (green) and inter-
stimulus intervals (ISIs)
was a high-contrast grating
of 8 s. TMS (gray box) was
applied during an ISI. TMS
pulse trains were varied in
(not shown) were recorded continuously; activity during TMS was not analyzed because of artifact
contamination (fig. S3A). (B) The full TMS trial. Evoked activity represents neural responses during stimulus
presentations, and spontaneous activity represents responses that occurred during ISIs.
Firing rate (spikes/s)
4 Hz, 2 s
Fig. 2. Effects of TMS on
imaging signals. Shown
are average time courses
of (A) spiking activity, (B)
LFP power, (C) tissue oxy-
gen, and (D) total hemo-
globin (Hbt) before and
after TMS (gray box). All
signals are expressed as a
percent change from their
pre-TMS baselines. Shaded
areas represent ±1 SEM.
(A) Spontaneous (left) and
evoked (right) spiking ac-
tivity (n = 47 cells). (B)
Spontaneous (left) and
evoked (right) LFP power
(n = 42 sites). (C) Tissue
oxygen (n = 21 sites). (D)
Hbt (n = 3 animals). Insets
in (C) and (D) show initial
increases. Time periods
containing TMS artifacts
were removed (fig. S3B). In (D), Hbt was measured by recording the change in 570-nm light reflectance
(∆R/R) from the cortical surface (upper right); scale bar, 1 mm.
∆ Firing rate (%)
n = 47 cells
Time (s)Time (s)
0 1002003000 100200300
∆ EVOKED LFP POWER
∆ SPONTANEOUS LFP POWER
Time (s)Time (s)
∆ Oxygen (%)
n = 21 sites
-∆ R/R (%)
n = 3 cats
Time delay (s)
-60 -300 3060
∆ Firing rate (spikes/s)
∆ Oxygen (%)
δ θ α β γ hγ
δ θ α β γ hγ
Fig. 3. Covariation between neural and oxygen
a function of TMS pulse frequency. Neural activity
was indexed by spontaneous spiking during the
initial phase (0 to 20 s after TMS) and by evoked
bars in this and subsequent panels represent ±1
SEM; where error bars are not visible, the error
was smaller than the plot symbol. (B) Time-lag
correlation between oxygen and neural signals
(left: spiking activity, n = 117 trials; right: LFP
power, 1 to 150 Hz, n = 77 trials). Positive time
lags indicate a shift of the neural signal forward in
time relative to the oxygen signal. Neural-oxygen
correlations were performed for evoked spiking
and LFP activity (green) and for spontaneous LFP
signals (purple); a similar analysis with spontane-
ousspikingcouldnot beperformedbecauseof low
the dashed lines are significant over the popula-
at positive time lags that are significantly greater
than those at negative delays (P < 0.05, paired
t test). (C) Neural-oxygen correlation magnitude
across bands. LFP bands are defined as follows: d
(delta; 1 to 4 Hz), q (theta; 4 to 8 Hz), a (alpha; 8
to 12 Hz), b (beta; 12 to 20 Hz), g (gamma; 20 to
80 Hz), hg (high-gamma; 80 to 150 Hz), all (1 to
150 Hz). (D) Neural-oxygen correlation latencies
VOL 317 28 SEPTEMBER 2007
higher pulse frequencies (Fig. 3A, lower left
quadrant). A more limited data set for pulse train
duration showed an analogous trend (fig. S7B).
These data suggest that the physiological effects
of TMS increase in a dose-dependent manner
within this regime of TMS application.
The relationship between decreases in oxy-
gen and neural activity is consistent with recent
studies of negative hemodynamic responses
(18, 19). However, reductions in oxygen may be
a cause of neural suppression rather than a con-
sequence. In this scenario, normal neural func-
tion would be limited by hypoxic conditions
(12). To investigate this possibility, we performed
a time-lag correlation analysis of simultaneously
acquired neural and oxygen data (fig. S8). Both
spike rate (Fig. 3B, left) and LFP power (Fig. 3B,
right) showed significant correlations with oxy-
gen across a broad range of time lags (P < 0.05, t
test). Notably, correlation coefficients were sig-
nificantly greater at time lags in which the neural
signal was shifted forward in time (P < 0.05,
paired t test). Furthermore, LFP-oxygen correla-
tions were band-specific with regard to magni-
tude and latency. Gamma and high-gamma bands
exhibited the strongest correlations (Fig. 3C), as
reported in previous studies of hemodynamic cou-
pling (20). Higher-frequency bands also exhib-
ited peak correlations at shorter latencies (Fig.
3D). This trend, which was most pronounced for
spontaneous activity, resulted from initial re-
sponse increases present in higher-frequency but
not lower-frequency bands (Fig. 2B, left). These
analyses, along with additional experiments (fig.
S9) (12), suggest that oxygen responses follow
neural activity in a manner consistent with neuro-
vascular coupling (21–24).
A striking aspect of our data is the long
duration of neural and hemodynamic changes
given the short application of TMS. Although
most human studies using similar stimulation
paradigms have demonstrated short-term effects,
several studies have noted changes in cortical
excitability on the order of minutes (25, 26).
Human studies using longer-duration stimula-
tion have shown effects lasting hours or even
days (27, 28). Such long-term changes in neural
function are thought to develop via spike
timing–dependent plasticity (27, 29, 30). Nota-
bly, alterations in synaptic efficacy have been
linked to changes in the temporal relationship
between spikes and LFP oscillations (31, 32).
To examine our data for a link between spike
timing and long-term neural changes, we per-
formed an additional analysis of phase rela-
tionships between single-unit spikes and LFP
oscillations (12, 33). For pre- and post-TMS time
windows, we quantified the degree of phase
locking from the distribution of LFP phases at
which spikes occurred (Fig. 4A). A striking ex-
ample of TMS-induced changes in phase distri-
butions is shown for spontaneous activity in Fig.
4B. Compared to the pre-TMS baseline (blue),
spike timing relative to the theta oscillation was
strongly desynchronized, as evidenced by the
increased spread of the distribution after TMS
(red). Across all frequency bands, spontaneous
activity showed significant reductions in phase
locking within the first 30 s after TMS (Fig. 4C,
left, P < 0.05, randomization test). By 90 s, this
index approached baseline values, and in the
gamma band it actually exhibited a significant
increase (P < 0.05). Somewhat similar effects
were present in evoked activity (Fig. 4C, right).
Phase locking to oscillations in the delta band
were strongly reduced, whereas increases were
present in both the gamma and high-gamma
precise timing of signals between interconnected
neurons advocates its ability to alter brain plas-
Consistent with previous work (29, 34, 35),
our results reveal long-lasting neural and hemo-
dynamic consequences of TMS that covary with
stimulation duration and frequency. In contrast,
other studies have reported a distinction where-
by low-frequency stimulation (≤1 Hz) causes
suppression and high-frequency stimulation
(≥8 Hz) leads to facilitation (5). However, this
division appears to be oversimplified (5, 36).
The precise effects of brain stimulation are fun-
damentally dependent on many factors (37).
For example, several groups have found that
identical TMS paradigms elicit opposite physi-
ological effects when applied to neighboring
cortical regions (34, 38) or different subjects
(36). Within a single site, TMS can produce dif-
ferential effects depending on the activity state
to which stimulation is paired (36, 39). Such
reports of variability and state-dependence reveal
the complex action of TMS, yet also hint at its
potential flexibility as an interventional technique.
Harnessing this potential requires the ability
to measure the precise neural effects of TMS
over different brain regions and time intervals.
Our findings show that TMS-induced modifica-
tions of neural activity are readily observed in
cerebral hemodynamics, which can be detected
by standard neuroimaging techniques. This re-
in which correlations were reported between
TMS-induced behavioral changes and hemo-
dynamic signals in functionally related brain
regions (39, 40). The capacity of brain imaging
to monitor the temporal progression of physio-
logical changes induced by TMS may prove
highly beneficial for the development and opti-
mization of both basic neuroscience and clinical
References and Notes
1. N. Singh, V. Pillay, Y. E. Choonara, Prog. Neurobiol. 81,
2. A. T. Barker, R. Jalinous, I. L. Freeston, Lancet i, 1106
3. A. Pascual-Leone, V. Walsh, J. Rothwell, Curr. Opin.
Neurobiol. 10, 232 (2000).
4. M. S. George, R. H. Belmaker, Transcranial Magnetic
Stimulation in Clinical Psychiatry (American Psychiatric
Publishing, Washington, DC, ed. 1, 2007).
5. P. B. Fitzgerald, S. Fountain, Z. J. Daskalakis,
Clin. Neurophysiol. 117, 2584 (2006).
6. A. T. Sack, D. E. Linden, Brain Res. Brain Res. Rev. 43, 41
7. Y. Zheng et al., Neuroimage 16, 617 (2002).
8. M. Suh, S. Bahar, A. D. Mehta, T. H. Schwartz, J. Neurosci.
25, 68 (2005).
9. T. Takano et al., Nat. Neurosci. 10, 754 (2007).
10. J. K. Thompson, M. R. Peterson, R. D. Freeman, Science
299, 1070 (2003).
11. The partial pressure of oxygen within extravascular tissue
is sensitive to changes in both cerebral blood flow and
the rate of oxidative metabolism (10).
12. See supporting material on Science Online.
13. Hbt is quantitatively related to cerebral blood flow (14)
and therefore provides an independent measurement of
14. R. L. Grubb Jr., M. E. Raichle, J. O. Eichling, M. M. Ter-
Pogossian, Stroke 5, 630 (1974).
15. During suppressive phases, we use evoked rather than
spontaneous spiking to index neural activity because
Fig. 4. Effects of TMS on
spike timing relative to LFP
oscillations. (A) Illustration of
phase locking between spikes
(red) and LFP (black). During
periods of high phase locking
(top), spikes occur at consist-
ent phases in the LFP (left),
and the resulting phase dis-
tribution is narrow (right).(B)
Example of a TMS-induced
change in phase locking. Be-
fore TMS (blue), spontaneous
oscillation. In the first 30 s
after TMS (red), the phase
distribution broadens, indicat-
ing a decrease in phase locking. (C) Changes in phase locking across LFP frequency bands for spontaneous
first 30 s after TMS; dark bars show changes at 60 to 90 s. Asterisks indicate significance (P < 0.05,
HIGH PHASE LOCKING
LOW PHASE LOCKING
LFP, 5-10 Hz
∆ Phase locking
28 SEPTEMBER 2007 VOL 317
baseline spontaneous activity (~1 spike/s) is typically too
low to detect signal decreases.
16. A. von Stein, J. Sarnthein, Int. J. Psychophysiol. 38, 301
17. One might expect to observe visually elicited oxygen
responses during the evoked intervals of the TMS trial
(fig. S2B). However, stimulus-evoked responses are
considerably smaller and more variable than TMS-
induced oxygen responses, and are therefore negligible
in the current paradigm (fig. S6).
18. A. Shmuel, M. Augath, A. Oeltermann, N. K. Logothetis,
Nat. Neurosci. 9, 569 (2006).
19. B. N. Pasley, B. A. Inglis, R. D. Freeman, Neuroimage 36,
20. J. Niessing et al., Science 309, 948 (2005).
21. On the basis of previous work in awake animals (22) and
the similarity of neurovascular organization (23, 24)
across the cortex, we expect that preserved coupling after
TMS will generalize to a broad range of cortical regions
and physiological states.
22. J. Berwick et al., J. Cereb. Blood Flow Metab. 22, 670 (2002).
23. C. Iadecola, Nat. Rev. Neurosci. 5, 347 (2004).
24. H. M. Duvernoy, S. Delon, J. L. Vannson, Brain Res. Bull.
7, 519 (1981).
25. A. Pascual-Leone, J. Valls-Sole, E. M. Wassermann,
M. Hallett, Brain 117, 847 (1994).
26. B. Takano et al., Neuroimage 23, 849 (2004).
27. Y. Z. Huang, M. J. Edwards, E. Rounis, K. P. Bhatia,
J. C. Rothwell, Neuron 45, 201 (2005).
28. F. Maeda, J. P. Keenan, J. M. Tormos, H. Topka,
A. Pascual-Leone, Clin. Neurophysiol. 111, 800 (2000).
29. H. Wang, X. Wang, H. Scheich, Neuroreport 7, 521
30. S. F. Cooke, T. V. Bliss, Brain 129, 1659 (2006).
31. C. Holscher, R. Anwyl, M. J. Rowan, J. Neurosci. 17, 6470
32. V. Wespatat, F. Tennigkeit, W. Singer, J. Neurosci. 24,
33. J. Jacobs, M. J. Kahana, A. D. Ekstrom, I. Fried,
J. Neurosci. 27, 3839 (2007).
34. T. Paus et al., J. Neurophysiol. 79, 1102 (1998).
35. H. R. Siebner et al., Neuroimage 14, 883 (2001).
36. A. Pascual-Leone et al., J. Clin. Neurophysiol. 15, 333
37. M. L. Kringelbach, N. Jenkinson, S. L. Owen, T. Z. Aziz,
Nat. Rev. Neurosci. 8, 623 (2007).
38. A. M. Speer et al., Biol. Psychiatry 54, 826 (2003).
39. A. T. Sack et al., Cereb. Cortex (2007).
40. C. C. Ruff et al., Curr. Biol. 16, 1479 (2006).
41. We thank our colleagues at the University of California,
Berkeley, and anonymous reviewers for their helpful
comments, and R. Bartholomew, N. Lines, and L. Gibson for
assistance in developing the electrophysiological apparatus.
Supported by research and CORE grants from the National
Eye Institute (EY01175 and EY03176, respectively) and by
NSF graduate research fellowship 2003014861.
Supporting Online Material
Materials and Methods
Figs. S1 to S9
13 June 2007; accepted 21 August 2007
Genomic Minimalism in the
Early Diverging Intestinal Parasite
Hilary G. Morrison,1* Andrew G. McArthur,1Frances D. Gillin,2Stephen B. Aley,3
Rodney D. Adam,4Gary J. Olsen,5Aaron A. Best,6W. Zacheus Cande,7Feng Chen,8
Michael J. Cipriano,1Barbara J. Davids,2Scott C. Dawson,9Heidi G. Elmendorf,10
Adrian B. Hehl,11Michael E. Holder,1Susan M. Huse,1Ulandt U. Kim,1Erica Lasek-Nesselquist,1
Gerard Manning,12Anuranjini Nigam,4Julie E. J. Nixon,1Daniel Palm,13
Nora E. Passamaneck,1Anjali Prabhu,4Claudia I. Reich,5David S. Reiner,2John Samuelson,14
Staffan G. Svard,15Mitchell L. Sogin1
The genome of the eukaryotic protist Giardia lamblia, an important human intestinal parasite, is
compact in structure and content, contains few introns or mitochondrial relics, and has simplified
machinery for DNA replication, transcription, RNA processing, and most metabolic pathways. Protein
kinases comprise the single largest protein class and reflect Giardia’s requirement for a complex
signal transduction network for coordinating differentiation. Lateral gene transfer from bacterial and
archaeal donors has shaped Giardia’s genome, and previously unknown gene families, for example,
cysteine-rich structural proteins, have been discovered. Unexpectedly, the genome shows little
evidence of heterozygosity, supporting recent speculations that this organism is sexual. This genome
sequence will not only be valuable for investigating the evolution of eukaryotes, but will also be
applied to the search for new therapeutics for this parasite.
its incidence may be as high as 0.7% (1). World-
wide, giardiasis is common among people with
poor fecal-oral hygiene, and major modes of
transmission include contaminated water sup-
plies or sexual activity. Flagellated giardial
trophozoites attach to epithelial cells of the small
intestine, where they can cause disease without
triggering a pronounced inflammatory response.
There are no known virulence factors or toxins,
and variable expression of surface proteins may
allow evasion of host immune responses and
ozoites can differentiate into infectious cysts that
are transmitted through feces.
Unusual features of this enigmatic protist in-
clude the presence of two similar, transcription-
iardia lamblia (syn. G. intestinalis, G.
duodenalis) is the most prevalent para-
sitic protist in the United States, where
ally active diploid nuclei and the absence of
mitochondria and peroxisomes. Giardia is a
member of the Diplomonadida, which includes
both free-living (e.g., Trepomonas) and parasitic
tify Giardia as a basal eukaryote (2–4). Other
gene trees position diplomonads as one of many
eukaryotic lineages that diverged nearly simulta-
neously with the opisthokonts and plants. Dis-
coveries of a mitochondrial-like cpn60 gene and
a mitosome imply that the absence of respiring
mitochondria in Giardia may reflect adaptation
to a microaerophilic life-style rather than diver-
drial ancestor (5, 6). Because of its impact on
humandisease andits relevance to understanding
the evolution of eukaryotes, we embarked upon a
genome analysis of G. lamblia.
The genome of G. lamblia WB clone C6
five chromosomes. The edited draft genome se-
quence contains 306 contigs on 92 scaffolds
(Supporting Online Material). The genome is
compact. We identified 6470 open reading
frames (ORFs) with a mean intergenic distance
of 372 base pairs (bp) (Table 1). Approximately
nucleotides (nt) of an adjacent ORF. Serial anal-
ysis of gene expression (SAGE) and cDNA se-
quences provided transcriptional evidence for
Although the total number of ORFs is similar
to that of yeast, many specific giardial pathways
appear simple in comparison with those of other
eukaryotic organisms. Giardia’s genome encodes
a simplified form of many cellular processes:
fewer and more basic subunits, incorporation of
single-domain bacterial- and archaeal-like en-
1Marine Biological Laboratory, Woods Hole, MA 02543–
Diseases, University of California, San Diego, CA 92103–
8416, USA.3Department of Biological Sciences, University
of Texas at El Paso, El Paso, TX 79968–0519, USA.4De-
partments of Medicine and Immunobiology, University of
Arizona College of Medicine, Tucson, AZ 85724–5049, USA.
5Department of Microbiology, University of Illinois at
Urbana-Champaign, Urbana, IL 61801, USA.6Department
of California, Berkeley, CA 94720–3200, USA.8University of
Pennsylvania, Philadelphia, PA 19194, USA.9University of
California, Davis, CA 95616, USA.
Georgetown University, Washington, DC 20057, USA.11In-
stitute of Parasitology, University of Zürich, CH-8057 Zürich,
Switzerland.12Razavi Newman Center for Bioinformatics, The
Salk Institute for Biological Studies, La Jolla, CA 92037–1099,
Institutefor InfectiousDisease Control, 171 82 Solna, Sweden.
USA.15Department of Cell and Molecular Biology, Uppsala
University, SE-751 24 Uppsala, Sweden.
*To whom correspondence should be addressed. E-mail:
13Centre for Microbiological Preparedness, Swedish
VOL 31728 SEPTEMBER 2007
28 SEPTEMBER 2007VOL 317SCIENCEwww.sciencemag.org
NEWS OF THE WEEK
In recent years, neuroscientists and psychia-
trists alike have touted the potential uses of a
noninvasive brain stimulation technique called
transcranial magnetic stimulation (TMS). The
method has been used to disrupt neural activity
experimentally in studies of human cognition,
and it has shown promise in clinical trials for
treating psychiatric disorders such as
depression (Science, 18 May 2001, p. 1284).
Although widely considered safe—thousands
of people have received TMS—relatively little
is known about how it actually works. Now, a
detailed look at its effects shows that TMS can
boost or dampen the firing of neurons depend-
ing on ongoing brain activity.
Neuroscientists at the University of Cali-
fornia, Berkeley, applied TMS to the cerebral
cortex of cats while monitoring neural activ-
ity and metabolism. Their findings, reported
on page 1918—and future investigations of
this type—will have important implications
for how TMS is used in people, other
One interesting possibility, according to
Mark George, a psychiatrist at the Medical
University of South Carolina in Charleston, is
that it may matter what subjects think about
while they’re being stimulated, a factor that
hasn’t received much consideration to date.
George, who pioneered TMS therapy for
depression, says a better understanding of how
TMS works will enable researchers and clini-
cians to apply it more effectively: “This is pre-
cisely where the field needs to go.”
In a typical TMS procedure, technicians
place a ring-shaped paddle near the scalp.
Electric currents swirling inside the paddle
produce a magnetic field that in turn gener-
ates currents in the underlying brain tissue.
These currents alter the electrical activity of
neurons, but exactly how they alter it is
Led by Ralph Freeman and graduate stu-
dents Elena Allen and Brian Pasley, the Berke-
ley scientists applied TMS to the visual cortex
of anesthetized cats and tracked the aftermath
using probes developed in Freeman’s lab that
can simultaneously record the electrical activ-
ity of neurons and measure fluctuations in oxy-
gen concentration, an indicator of energy con-
sumption. Using optical imaging methods, the
researchers also tracked hemoglobin levels,
another metabolic marker. A train of TMS
pulses lasting a few seconds caused an imme-
diate increase in neural firing that lasted for
about a minute, followed by a decrease in firing
for several minutes. Oxygen and hemoglobin
mirrored this pattern, indicating that neurons’
firing and energy demands go hand in hand.
TMS had a dramatically different effect,
however, on neural activity evoked by black
and white bars flashed on a computer
screen. (Such responses persist even in anes-
thetized animals.) In this case, neural firing
dipped sharply after TMS and remained
suppressed for several minutes.
The findings have implications for design-
ing TMS therapies, says George. For depres-
sion therapy, for example, “we may need peo-
ple to become sad in the chair while stimulat-
ing [them],” George says. “Alternatively, we
might have them engage in formal cognitive
therapy, thinking positive thoughts.” Such con-
siderations are important, he adds, as the Food
and Drug Administration is considering
approval for daily TMS of the prefrontal cortex
to treat depression.
The new findings also suggest why the
effects of TMS often vary, says Alvaro Pascual-
Leone, a neurologist at Harvard Medical
School in Boston. Pascual-Leone suggests that
TMS results could be made more consistent
by monitoring the physiological state of the
brain using electroencephalography or functional
magnetic resonance imaging. –GREG MILLER
Uncovering the Magic in Magnetic Brain Stimulation
Pollution Slows China’s Canal Project
The first phase of a massive project to
replumb some of China’s mightiest water-
ways has fallen far behind schedule because
local authorities don’t want to pay for the priv-
ilege of drinking polluted water.
The South-to-North Water Diversion
Project is a three-stage effort to alleviate
chronic water shortages in the country’s
more populous but parched northern plains
(Science, 25 August 2006, p. 1034). The
eastern route makes use of an existing net-
work of canals, rivers, and lakes to pump and
move water from the lower Yangtze River to
Jiangsu and Shandong provinces. But this
month, the official Xinhua news agency
announced that the first phase of the route,
scheduled to begin operating this year, has
been delayed at least 3 years.
Nearly half of the $4 billion cost of the
first phase is earmarked for improving the
quality of the water. However, the central
government is footing only about 10% of the
bill, with the rest expected to come from
localities that will benefit from the project.
But because nobody wants to clean up some-
body else’s dirty water, few treatment facili-
ties have been built along the route, and water
quality continues to deteriorate. So far this
year, according to Xinhua, the water is drink-
able at only one of the 21 monitored cross
sections in Shandong.
Some engineering experts say the entire
project itself needs to be rethought, with a
greater emphasis placed on improving the
ecology of the Yellow and Huai river basins.
Dredging the Grand Canal north of the Yellow
River to make the ancient waterway navigable,
they say, would provide a greater benefit to the
region and, thus, attract more investment.
Qian Ye, a climate researcher at the
U.S. National Center for Atmospheric
Research in Boulder, Colorado, thinks the
Chinese government should do a more com-
prehensive feasibility study of the project that
considers the impact of climate change.
Global warming, he says, could make China’s
north wetter and allow authorities to scale
back the controversial project. –HAO XIN
CREDIT: COURTESY OF HANNA MÄKI/LABORATORY OF BIOMEDICAL ENGINEERING, HELSINKI UNIVERSITY OF TECHNOLOGY AND BIOMAG LABORATORY, HELSINKI UNIVERSITY HOSPITAL
Stimulating results. New research hints at the
mechanisms of magnetic brain stimulation.
Published by AAAS
Supporting Online Material for
Transcranial Magnetic Stimulation Elicits Coupled Neural and Hemodynamic
Elena A. Allen, Brian N. Pasley, Thang D. Duong, Ralph D. Freeman*
*To whom correspondence should be addressed. E-mail: email@example.com
Published 28 September 2007, Science 317, 1918 (2007)
This PDF file includes:
Materials and Methods
Figs. S1 to S9
Neural and hemodynamic effects of TMS
Allen et al
- 1 -
Supporting Online Material
Materials and Methods:
A total of eight mature cats were used in this study. All procedures complied with the National
Institutes of Health Guide for the Care and Use of Laboratory Animals and were approved by the
Animal Care and Use Committee at the University of California Berkeley.
Surgery and Anesthesia: Surgical anesthesia was induced with isofluorane (4%). After
placement of venous catheters, anesthesia was continued with intravenous infusion of fentanyl
citrate (10 μg.kg-1.hr-1) and thiopental sodium (6.0 mg. kg-1.hr-1). Additional bolus injections of
thiopental sodium were given as needed during surgery. Following the placement of a tracheal
cannula, animals were artificially ventilated with a 25% O2 / 75% N2O mixture. Respiration rate
was adjusted to maintain expired CO2 between 30-36 mmHg (generally between 15-25
breaths/min). In one cat, moderate hypercapnia was induced to manipulate levels of tissue
oxygen. For this procedure, 5% CO2 was added to the inspired gas, and the fraction of N2O was
reduced to keep inspired O2 constant. After 4.5 minutes, the gas mixture was returned to 25% O2
/ 75% N2O. A minimum of 30 minutes was provided for recovery before beginning another
period of hypercapnia. Body temperature was maintained at 38º C with a closed-loop controlled
heating pad (Love Controls, IN, USA). A craniotomy was performed over area 17 (Horsley-
Clarke coordinates P4, L2 (1)) and the dura was resected. The craniotomy was then covered
with agar and wax to form a closed chamber. After completion of surgical procedures, fentanyl
citrate infusion was discontinued and the rate of thiopental sodium infusion was gradually
lowered to a level at which the animal was stabilized. This level of infusion was determined
individually for each animal (typically 1.5 to 2.0 mg. kg-1.hr-1). After stabilization, the animal
was immobilized with an intravenous infusion of pancuronium bromide (0.2 mg.kg-1.hr-1) to
prevent eye movements. The pupils were dilated with drops of 1% atropine sulfate, and
nictitating membranes were retracted with 2.5% phenylephrine hydrochloride. Rigid contact
lenses with 3 mm artificial pupils were fitted to the eyes. EEG, ECG, heart rate, temperature,
end-tidal CO2, and intra-tracheal pressure were monitored through the duration of the
Visual Stimulation: Visual stimuli consisted of sinusoidal gratings presented on a luminance-
calibrated CRT monitor (85 Hz refresh rate, mean luminance 45 cd/m2). Preliminary tests were
performed on each neuron to identify the stimulus orientation, spatial frequency, temporal
frequency, position, and size to maximize the neuron's spike response. During TMS trials
drifting gratings with optimal parameters were displayed at 50% contrast for 2 s.
TMS: TMS was applied to the cat visual cortex using a MagStim Rapid system (The MagStim
Company Ltd., Whitland, UK) with a 70 mm figure-eight coil. The coil was positioned using a
mechanical camera arm. For most experiments the coil was positioned posterior to the visual
cortex and angled 45 degrees towards the horizontal plane, which permitted an electrode
penetration into area 17 at Horsely-Clarke coordinates P3.5, L2, with an angle of P45, M0 (fig.
S1A). During optical imaging experiments, the coil position was adjusted to create an
unobstructed view for the macrolens. This involved reducing the angle between the face of the
TMS coil and the coronal plane to 10 to 20 degrees (fig. S1B). In experiments using tungsten
electrodes, the coil was positioned obliquely near the transverse plane superior to the visual
cortex, and the angle of the electrode penetration was instead made at P4.5,L2, with an angle of
Neural and hemodynamic effects of TMS
Allen et al
- 2 -
A45, M0 (fig. S1C). In all configurations, the midpoint of the coil was centered on the
craniotomy in area 17 and was located between 1 and 2 cm from the skull. Pulse trains were
delivered by TTL digital pulses with parametrically varying frequency (1,4,8 Hz) and duration
(1,2,4 s) at 100% stimulation intensity. At this intensity and range of distances (1-2 cm distance
from the skull), the induced electric field strength is estimated to be ~100-200 V/m (2). Given
that magnetic stimulation induces an electric field which is relatively insensitive to the
conductivity of the skull and protective tissues, the small (~2-4 mm) craniotomy and duratomy
used in our preparation are not expected to significantly alter induced currents (3). In
experiments that varied TMS pulse train duration and frequency, parameters were varied
randomly across trials. To ensure sufficient neural recovery after each TMS trial, stimuli were
displayed periodically until the evoked response maintained a steady state value for over 1
minute. A minimum of 6 minutes was provided between TMS applications, and 10-15 minute
intervals were typical.
Electrophysiology: Simultaneous oxygen and neural measurements were made using multibarrel
carbon fiber microelectrodes (CARBOSTAR-7, Kation Scientific, Minneapolis, MN). These
electrodes are composed of a central glass barrel containing a spark-etched carbon fiber,
surrounded by 7 pipette barrels (4). The total tip diameter is between 10 and 20 μm, and the
conical carbon tip extends roughly 15 μm beyond the glass pipette tips (4). The carbon fiber was
used to measure tissue oxygen levels (as described below), and extracellular neural recordings
were made by filling one of the surrounding barrels with 3 M NaCl (8-10 MΩ at 1KHz). Neural
signals passed first through a high impedance head-stage, then through a preamplifier (Plexon,
Dallas TX). At this stage, the signal was amplified and separated into two components. The
LFP signal was obtained via analog filtering of the broadband neural signal with a low-pass
cutoff of 0.7 Hz and a high-pass cutoff of 170 Hz. The LFP signal was then digitized at 500 Hz.
The multi-unit signal was obtained by filtering between 500 Hz and 8 MHz. From this signal,
single units were discriminated online based on the temporal shapes of their extracellular
potentials. Spike times were recorded with 0.04 ms precision.
Carbon fibers were converted into amperometric oxygen sensors by applying a negative
polarization voltage between the carbon tip (cathode) and an external Ag/AgCl reference
electrode (anode) (5). Amperometric oxygen sensors composed of carbon have been used
previously to measure oxygen tension in various biological tissues including brain tissue (6, 7).
Polarization between the anode and cathode, between -0.8 and -0.95 V, was set to the center of
the plateau region in the current voltage curve. Within this region, the sensor current is
determined by the diffusion of oxygen at the cathode and is relatively insensitive to small
changes in the applied voltage. A high impedance picoammeter (Unisense, Denmark) provided
the polarization voltage and measured the small currents generated by the sensor.
Calibration of the carbon fiber oxygen sensor was performed in a 0.9% NaCl solution.
The partial pressure of oxygen in the solution was varied by bubbling a mixture of oxygen and
nitrogen gas through the saline. Calibration was performed at a minimum of five levels of
oxygen, and a linear relationship between current and oxygen partial pressure was observed in all
electrodes tested (fig. S2A). To avoid possible deposition of biological material on the cathode
surface due to long term use (5), a new carbon fiber sensor was used in each experiment.
While it is possible that electroactive species present in brain tissue could affect oxygen
measurements, our data indicate that these contributions are minimal. We compared signals
Neural and hemodynamic effects of TMS
Allen et al
- 3 -
from the carbon fiber sensor with those from a Clark-style oxygen electrode (Unisense,
Denmark), which incorporates an oxygen selective membrane and guard cathode to increase the
sensitivity and precision of measurements. In in vivo tests, both electrodes were lowered into
visual cortex and were co-localized to within 100 μm. The simultaneously recorded oxygen
measurements were virtually equivalent, with correlation coefficients ranging between 0.9 and
0.99 across sites.
Prior to application of TMS, stimulus evoked oxygen responses were obtained at each
site. This protocol consisted of 15 to 60 repeated displays of a 4 s drifting grating, with a
variable inter-stimulus interval of 24-30 s. The average evoked oxygen responses obtained at all
sites were similar to those reported in previous studies (fig. S2B) (8).
In addition to neural-oxygen recordings with carbon fiber electrodes, we applied the same
TMS protocol to two animals in which single-unit recordings were made with epoxy coated
tungsten microelectrodes (5 MΩ). Tungsten electrodes were mounted in a dual array, allowing
us to efficiently increase the sample size of neural data. These experiments also allowed us to
confirm that the observed neural effects of TMS were independent of the type of recording
electrode and the exact coil configuration (fig. S1). No qualitative differences were found
between the neural responses recorded with carbon fiber electrodes and those recorded with
Optical Imaging: In optical imaging experiments (n = 3 mature cats), the brain was stabilized
with agar (1.5%) and images were obtained through a glass coverslip. The cortex was
illuminated with green light (570 ± 5 nm) from two fiber-optic light guides. At this wavelength
of illumination, oxygenated hemoglobin and deoxygenated hemoglobin have equal absorption
coefficents, and therefore changes in reflectance correspond to changes in the total level of
hemoglobin (Hbt) (9, 10). The reflectance signal was captured with a Dalsa CCD camera (Dalsa
INC., Ontario, Canada ) using a modified macroscope (2 Nikkor, 50-mm camera lenses mounted
front-to-front)(11). Images were stored at an effective frame rate of 6.75 Hz, using custom
software written by TD.
Before applying the TMS protocol, preliminary experiments were performed to ensure an
Hbt response to visual stimulation. Full-field drifting gratings at 50% contrast were displayed
for 4 s with a variable inter-stimulus interval of 20-30 s. In all experiments, the average change
in reflectance to 10 repetitions yielded a typical stimulus evoked hemodynamic response with a
peak of ~1 to 5% (10, 12). The TMS protocol during optical imaging trials was identical to that
during neural/oxygen trials except that no visual stimulation was presented. Initial
electrophysiology experiments demonstrated that tissue oxygen exhibited similar response
profiles to TMS regardless of visual stimulation (fig. S6). This was confirmed with statistical
analysis of the oxygen signal showing no response difference between evoked and spontaneous
time intervals (see Statistical Analysis). In addition, TMS was applied only at 4 or 8 Hz
frequency and 2 or 4 s duration.
All data analyses were performed in MATLAB (MathWorks, Inc, Natick, MA,. USA) and C++.
Neural: All data included in analyses were free of TMS-induced electrical artifacts. This was
achieved by excluding any neural data that occurred between TMS pulses or earlier than 100 ms
after the last pulse (fig. S3A)
Neural and hemodynamic effects of TMS
Allen et al
- 4 -
Single unit data were converted into spike rates by dividing the number of spikes in a
time window by the duration of that window. Note that the evoked activity was defined in terms
of the absolute spike rate (i.e., without subtraction of the pre-stimulus spontaneous firing rate).
This ensured that observed changes in evoked activity were not simply artifacts of changes in
spontaneous firing rates. For group effects, data were averaged together first by site (which
included trials with different TMS application parameters), then over sites to remove any biases
from sites with more trials. Changes in spike rate were converted into percent changes from
baseline for plotting purposes.
Changes in LFP power were computed in three steps. First, the continuous LFP signal
for each trial was divided into smaller 2 and 8 s segments corresponding to the stimulus and
inter-stimulus intervals. Line noise at 60 Hz and 85 Hz (monitor refresh rate) was removed by
subtracting the best-fit continuous sine wave from the raw signal (http://chronux.org). Second,
LFP spectrograms were computed using multi-taper spectral estimation (13, 14). The power
spectrum was estimated on a 1 s window with 5 Hz bandwidth using nine Slepian data tapers.
Third, the power at each frequency and time bin was normalized by the average power for that
frequency in the baseline period (40 s). Note that this conversion was specific to the type of LFP
signal, i.e., stimulus-evoked LFP power was normalized by the baseline power during stimulus
presentations, and spontaneous LFP power was normalized by the baseline power during inter-
stimulus intervals. Therefore, modulations in power represent TMS-induced, rather than
stimulus-induced changes. Modulation in LFP power was first averaged over trials within site,
and site-specific responses were subsequently averaged together to obtain the population
responses. LFP power changes were converted into percent changes from baseline for plotting
Oxygen: Because the precise relationship between oxygen partial pressure and sensor current
can change with temperature, pressure, and the diffusion properties of the solution in which
measurements are made (5), no attempt was made to measure oxygen tension in absolute terms.
Oxygen signals were converted to a percent change from the average signal during the baseline
period (40 s). To examine the population time course, trials were averaged within the same site,
then over all sites.
TMS application produced transient disruptions in the tissue oxygen signal which lasted
1-3 s beyond the cessation of the pulse train (fig. S3B). These periods were excluded from all
analyses of tissue oxygen. Note that this artifact prevented analysis of the early time period
which could potentially include an initial dip in the oxygen response (8). Previous studies have
demonstrated an initial negative deflection of the oxygen signal that is closely linked to local
neural activity and presumably reflects an increase in oxygen consumption (8). In accordance
with these earlier studies, our preliminary optical imaging experiments (which do not contain
TMS artifacts, see Optical Imaging) have revealed significant increases in deoxyhemoglobin in
the first 4 s post- TMS (data not shown). These data are consistent with an initial increase in
oxidative metabolism due to neural activation following TMS.
Optical Imaging: Regions of interest were manually defined and typically included the full
extent of exposed cortical surface. Manual delineation and an intensity threshold were used to
exclude pixels corresponding to large vessels in order to target responses in the
microvasculature. For each pixel, linear trend was removed and the baseline was calculated as
the mean intensity of all time points preceding the TMS pulse train. The time course of each
Neural and hemodynamic effects of TMS
Allen et al
- 5 -
pixel was then converted to a percent signal change by subtracting and then dividing each time
point by the baseline value. Finally, to generate the mean Hbt time course, the resulting time
series were averaged over all pixels, repetitions, and experiments.
Due to the nature of the optical signal, no TMS artifacts were observed. This was verified
with recordings of a static, non-physiological test surface. Variations in the reflectance signal
during TMS application were within noise levels of the camera (data not shown).
Statistical Analysis: Separate 3-way repeated measures ANOVAs were conducted for the spike
and oxygen data. The ANOVAs included 3 factors: activity state (evoked vs. spontaneous),
TMS pulse frequency (1, 4, or 8 Hz), and time (0-20 s post-TMS, 30-90 s, 180-210 s), with time
as the repeated measure. Due to the limited number of trials over which pulse duration was
varied, it was not incorporated as a factor. The ANOVAs included only trials with 4 s pulse
duration, as they constituted the largest proportion of the data set. For oxygen data, the first time
interval was 8-20 s to exclude the TMS artifact.
In post-hoc analyses of oxygen and spiking data, statistical significance was assessed by
pairwise comparisons of signals before and after TMS for each trial (Wilcoxon sign rank tests).
Identical tests were applied to the additional LFP and Hbt data sets. Unless otherwise noted, all
P-values were Bonferroni corrected for multiple comparisons over time intervals.
Time-Lag Correlation Analysis: Time-lag correlations were performed between simultaneously
collected neural and oxygen trials (fig. S8). A correlation analysis between spontaneous spiking
activity and tissue oxygen could not be performed due to low firing rates (typically less than 1
spike/s during baseline) and the rectifying nonlinearity of spiking activity (i.e., spike rate cannot
become negative). Neural signals were shifted from -60 to +60 s in 5 s steps. Positive delays
represent a shift in the neural signal forward in time (rightward) relative to the oxygen signal.
Pearson’s correlation coefficients were computed between the raw neural response level and the
raw oxygen level averaged over a 5 s window. Correlation coefficients were converted into z-
scores using Fisher's Z-transform, which reduces skew and approximates a normal distribution
(15). Mean z-scores were reverse transformed into correlation coefficients for plotting purposes.
Correlations at individual time-lags were considered positive when the mean of the z-score
distribution was significantly larger than zero (one-tailed t-test). Z-scores at different time-lags
were compared using a paired t-test.
Time-lag correlation curves for different neural bands were compared with regard to their
magnitudes and shapes. For each curve, the correlation magnitude was estimated by averaging
the z-scored correlation coefficient over time lags from +5 to +40 s. This time window was
defined to include time delays which showed significant correlations for all neural bands. To
assess differences in curve shape between neural bands, we calculated the center of mass of each
curve after it was normalized from 0 to 1. This metric of correlation latency provides a robust
estimate of the skew of the curve, without an implicit assumption of curve shape.
Analysis of Parametric TMS Application: Analysis of neural and oxygen changes to different
levels of TMS application included all trials with appropriate stimulation parameters.
Alterations in neural activity were calculated as a change in spike rate, rather than a percentage
change, due to the large difference between evoked and spontaneous baselines (e.g., a change of
1 spike/s is a ~100% change in spontaneous rate, yet only a ~3% change in evoked activity).
Neural and hemodynamic effects of TMS
Allen et al
- 6 -
Changes in spiking activity and tissue oxygen were calculated over identical time windows
(early changes: 0 to 20 s; late changes: 30 to 90 s post-TMS).
The approximately linear relationship between neural and oxygen signals is consistent
with current models of neurovascular coupling. It is also possible that components of the
hemodynamic response may be due to direct action of TMS on vascular tissue. However,
previous studies indicate that the excitation threshold of neural tissues is an order of magnitude
less than that of other tissues, including vascular smooth muscle cells (16-18). In addition,
application of tetrodotoxin to eliminate neural activation also abolished vascular responses to
direct electrical stimulation at intensities exceeding those used here (19). We therefore expect
that the observed hemodynamic changes are primarily of neural origin.
Phase Locking Analysis: A phase locking index was computed for trials in which we could
obtain artifact-free recordings of both isolated spikes from single units and LFPs from the same
electrode. This included 129 trials across 4 animals. We chose an index of phase locking that
utilizes the analytic signal approach to estimate the instantaneous LFP phase at which each spike
occurs (20). The phase locking index corresponds to one minus the circular variance of this
phase distribution (21, 22), which is also equivalent to the mean resultant vector strength (VS) in
directional statistics (23). We chose this measure of phase locking because of its relative
insensitivity to changes in signal amplitude. This is in contrast to alternative indices of
synchrony, such as coherence (14), that confound changes in signal amplitude with changes in
signal phase (20). In general, most analyses involve data characterized by amplitude changes,
and this is especially true in our data set which reveals large changes in both LFP power and
spike rate. In the analytic signal approach, phase information is retained while amplitude
information is discarded, so the phase locking index is derived independently from LFP
amplitude. This point is true as long as the signal-to-noise ratio (SNR) of the LFP is large
enough to display oscillatory behavior from which the instantaneous phase can be estimated. To
check this, we verified that analyses with different SNR cutoffs yielded equivalent results. In
addition, changes in spike rate might also influence this index due to statistical bias in the
circular variance. This possibility is corrected for by a spike count matching procedure
Data analysis proceeded as follows. After removal of line noise, the LFP trace for each
TMS trial was divided into data segments that corresponded to each 2 sec stimulus presentation
for evoked activity or each 8 sec ISI for spontaneous activity. These data segments were then
filtered in 5 Hz bands from 1-150 Hz using the EEGLAB (24) function ‘eegfilt’. This function
utilizes the MATLAB function ‘filtfilt’ which preserves phase information. Instantaneous
phases for these filtered segments were then extracted from the analytic signal obtained through
the Hilbert transform (20). Each spike was then assigned the phase value with the matching time
Because the circular variance statistic is biased by the number of observations (25) and
because TMS altered spike rate, it was necessary to take this change into account. Intuitively,
this problem can be understood in the following way. Increasing the number of spikes per
oscillation necessarily increases the spread in the phase distribution because some spikes must
now occur at different, additional phase values. With only a small number of spikes, these
additional phase values will not necessarily occur, and the variance of the phase distribution can
Neural and hemodynamic effects of TMS
Allen et al
- 7 -
To correct for the change in spike rate after TMS, we used the following matching
procedure. First, the spike count during each cycle of the LFP oscillation was tabulated for both
pre-TMS and post-TMS intervals. Second, VS was calculated separately for each spike count by
grouping cycles with matching numbers of spikes. Third, a set of ∆VS values was obtained by
subtracting pre-TMS VS values from the post-TMS VS values that were matched for spike
count. Fourth, the mean of these values over all spike counts yielded the final ∆VS. In order to
increase the number of cycles matched for spike count, data were collapsed into 30 second
periods (corresponding to 3 evoked/spontaneous intervals). This matching procedure ensured
that changes in VS after TMS could not be accounted for by changes in spike rate. This
procedure was applied to individual TMS trials which were then averaged to yield the population
average in Figure 4C. To calculate ∆VS for gamma (20-80 Hz) and high gamma (80-150 Hz)
bands in Figure 4C, values were averaged across 5 Hz frequency bands. Statistical significance
was assessed using a randomization test. The pre-TMS and post-TMS phases were shuffled and
∆VS was recalculated 5000 times to generate a null distribution. The actual ∆VS value was
deemed significant if it exceeded 95% (P < 0.05) of this null distribution after Bonferroni
correction for 24 multiple comparisons (i.e., 6 LFP bands X 2 time intervals X 2 stimulus
conditions). The standard errors in Figure 4C were estimated with a bootstrap procedure in
which a resampled distribution of ∆VS values was constructed by resampling the phases for each
time interval and 5 Hz frequency band 1000 times with replacement. The bootstrapped standard
error of ∆VS for each band and time point was computed as the standard deviation of the
resampled distribution. The standard errors for ∆VS in gamma and high gamma bands in Figure
4C were propagated from those of the individual 5 Hz bands across which each range was
Spontaneous vs. Evoked Activity: Intermittent presentation of visual stimuli allowed us to
examine how TMS affects both ongoing spontaneous activity and stimulus evoked activity. Our
results demonstrate robust differences between activity states (Fig. 2A, fig. S5). While the
detailed cellular and circuit mechanisms underlying these differences remain to be determined,
we can identify several mechanisms that are likely to contribute. For instance, the immediate
increase in spontaneous activity following TMS is presumably due to direct depolarization from
the induced electric field (26). Depolarization likely originates in a distinct population of
neurons (26) and may be amplified via recurrent synaptic excitation to generate spontaneous
activity in a larger population that outlasts the period of TMS application. Indeed, previous
studies using electrical stimulation have reported that single pulses elicit reverberating volleys of
In contrast to spontaneous activity, stimulus evoked activity depends on the intact
function of a cortical circuit. TMS may act through several intrinsic and extrinsic factors, such
as a predominance of inhibitory activation (28), intrinsically-mediated changes in polarization
(29), or short term depression (30) to mediate disruption of this function. Support for this
interpretation comes from our analysis of phase locking (Fig. 4) which demonstrates substantial
desynchronization of signals between interconnected neurons. The temporal disruption of
signals potentially interferes with the stimulus processing, thereby reducing the amplitude of the
Interestingly, recent human studies have shown somewhat similar state-dependent effects
(e.g., 31, 32). In these experiments, the application of TMS is paired to a specific brain state
(e.g., applied during a resting condition or, alternatively, during the execution of a cognitive
Neural and hemodynamic effects of TMS
Allen et al
- 8 -
task), and the resulting effects of TMS are dependent on this pairing. This paradigm differs from
the current study in that we applied TMS only during the inter-stimulus interval (i.e.,
spontaneous activity) and never with presentation of the visual stimulus. However, the
differential effect of TMS on spontaneous vs. evoked activity observed in the current study
suggests that further studies are warranted to clarify the influence of activity state on the
physiological response to TMS.