For Peer Review - Imperial College London...The paper “Peripheral inflammation alters endocannabinoid-induced modulation of nociceptive spinal dorsal horn synaptic transmission”

For Peer Review

Peripheral inflammation alters N-

arachidonoylphosphatidylethanolamine (20:4-NAPE)

induced modulation of nociceptive spinal cord synaptic

transmission

Journal: British Journal of Pharmacology

Manuscript ID 2017-BJP-0087-RPT-G.R1

Manuscript Type: Research Paper Themed Issue

Date Submitted by the Author: n/a

Complete List of Authors: Nerandzic, Vladimir; Institute of Physiology vvi CAS Mrozkova, Petra; Institute of Physiology vvi CAS Adamek, Pavel; Institute of Physiology vvi CAS Spricarova, Diana; Institute of Physiology vvi CAS Nagy, Istvan; Imperial College London, APMIC Palecek, Jiri; Institute of Physiology vvi CAS

Major area of pharmacology: Pain

Cross-cutting area: Inflammation

Additional area(s): Cannabinoid, Ion channels, Ligand-gated ion channels, Plasticity, TRP

British Pharmacological Society

British Journal of Pharmacology

For Peer Review

1

We would like to thank all the reviewers for their valuable comments. We have incorporated virtually

all of their suggestions into the revised version of the paper, while complying with the available strict

limits on the word count of the paper. We believe that the changes have significantly improved this

manuscript and we are grateful for all the reviewers help to achieve this improvement. We hope that

the reviewers as well as the editor will find the current version of our manuscript acceptable for

publication in the British Journal of Pharmacology.

In addition to making changes to the text, as was suggested, we have also added new data on

anandamide release from spinal cord slices in vitro under similar conditions to our

electrophysiological recordings. We assume that this new evidence of anandamide production after

20:4-NAPE bath application will further strengthen our experimental data and hypothesis.

Please find below our reply to the reviewers comments. Appropriate changes are marked in the text.

Editor

The present student provides useful data that shows that the putative amandamide precursor 20:4

NAPE acts to inhibit excitatory transmission in superficial dorsal horn neurons. The study also adds

to the literature by describing differential effects of CB1 and TRPV1 antagonists on NAPE-induced

inhibition in control vs carrageenan–induced inflammatory conditions. Whilst in general this is a

well-designed pharmacological study, there are a number of concerns that should be addressed.

Major points

The authors justify their use of 20:4-NAPE (vs amandamide) nicely at the very end of the

Discussion, but this should come much earlier in the paper;

We agree. We have changed the last paragraph of the abstract and the introduction to better reflect

the purpose of the study.

I also believe that this is the first study that investigated the electrophysiological action of 20:4-

NAPE in a model of inflammatory pain? If so, this should be highlighted.

We agree. To the best of our knowledge, the only available data on the electrophysiological action of

20:4-NAPE is from Varga and colleagues (2014) who have studied the effect of 20:4-NAPE on “naïve”

cultured primary sensory neurons. We have highlighted our novel investigation of 20:4-NAPE

application on spinal synaptic transmission in control and inflammatory conditions in the first

paragraph of the abstract.

Connected to this point, the tendency to ascribe 20:4-NAPE effects definitively to amandamide

should be tempered, as this remains unproven here.

We agree. We have added new data showing increased anandamide production under our

experimental conditions. These results strongly support our hypothesis that anandamide plays a key

role in the changes we observe after 20:4-NAPE application. There is also published evidence that

20:4-NAPE bath application leads to anandamide production from experiments with DRG cultures

(Varga et al. 2014). Although our additional experiments strongly support our original view, we have

changed some of the wording in the text in order to diminish the role of anandamide.

Page 1 of 76



For Peer Review

2

It would be useful to comment on what (if anything) is known about NAPE binding to CB1 and

TRPV1 and the availability of NAPE inhibitors to further interrogate pathways and mechanisms of

action.

We agree. We have no evidence available at this moment that would suggest or contradict direct

effect of 20:4-NAPE on the TRPV1 and CB1 receptors. Only circumstantial evidence is available on the

absence of direct effect of 20:4-NAPE on the TRPV1 receptor. 20:4-NAPE did not activate TRPV1

receptors in DRG cultures at room temperature, in contrast to responses evoked by capsaicin

application (Varga et al. 2014).

Overall, it is better the change the title to “Peripheral inflammation alters N-arachidonoyl

phosphatidylethanolamine (NAPE)-induced modulation of nociceptive spinal cord dorsal horn

synaptic transmission” and then present arguments in the Discussion as to why amandamide MAY

mediate effects; such arguments should include a more fuller description of evidence that NAPE

represents the major precursor for amandamide (Lu & Mackie, 2016, but also Wang J1, Ueda N.

Biology of endocannabinoid synthesis system. Prostaglandins Other Lipid Mediat. 2009).

We agree. Thank you for the suggestion. The title was changed accordingly and to reflect the

available character limit. New evidence was also added showing anandamide synthesis from 20:4-

NAPE in spinal cord slices. The suggested citations were added to the text.

Is there a compelling reason why miniature EPSCs were not recorded (such as very low event

frequency); similarly, the decision to focus on excitatory vs inhibitory transmission should be

justified.

Both excitatory and inhibitory currents are important for nociceptive signaling modulation at spinal

cord level. However, the excitatory signaling is at the end crucial for the transmission into the higher

brain centers. Also TRPV1 receptors at the spinal cord level are known to be predominantly expressed

on the presynaptic endings of excitatory primary afferents. Their presence and function at

postsynaptic neurons was highly disputed and there is only highly limited supportive evidence (see

Spicarova et al 2014 for review). In this paper we have thus concentrated on the NAPE effects on

excitatory synaptic transmission. We may study the effects on inhibitory transmission in another

study as it requires a completely new set of experiments. We have some preliminary data on the

effect of 20:4-NAPE application on the mEPSC properties that are in general agreement with the

sEPSC and eEPSC data presented here. As mEPSC currents recorded in the presence of TTX reflect

more of a modulatory effect in the synaptic transmission, we feel that it requires a separate paper to

fully reflect the findings.

Discussion on the clinical relevance of NAPE should be improved.

We agree. However, we feel that our electrophysiological experiments do not provide strong enough

evidence for a clear clinical significance. However, we have added some views of that in the abstract,

and in the end of discussion.

Minor points

Controls for carrageenan injection should be specified (naïve, sham or vehicle injection?)

The peripheral inflammatory pain model is well established. We have therefore found appropriate to

use naïve animals as controls. This is now clearly identified in the text.

Page 2 of 76



For Peer Review

3

Authors should better justify why a single concentration of 20 microM NAPE was used here.

We agree. The concentration of 20:4-NAPE used was based on data from experiments on DRG

cultures and also from our preliminary data. With the long experimental protocol used while

recording from visually identified neurons in spinal cord slices, it would prove very difficult to do any

experiments on concentration-response relationship with all the pharmacological treatments. Also

our experiments on anandamide release provided further support for the use of the 20 uM

concentration.

It would be useful to justify the use of PF514273 and SB366791 and it is unclear where capsaicin

(stated in Methods) was used in this study.

We agree. A text was added to the methods concerning the antagonists used and the capsaicin

treatment.

For all studies, authors must include Tables of Links to the Concise Guide to Pharmacology and

ensure that the correct nomenclature is stated and is used throughout the manuscript: see the BJP

editorial 'BJP is linking its articles to the IUPHAR/BPS Guide to PHARMACOLOGY'

(http://onlinelibrary.wiley.com/doi/10.1111/bph.13112/epdf).

We agree. The appropriate links were added.

Reviewers' Comments to Author:

Reviewer: 1

The paper “Peripheral inflammation alters endocannabinoid-induced modulation of nociceptive

spinal dorsal horn synaptic transmission” deals with the effect of an anandamide precursor, the

20:4-NAPE, on the spontaneous and evoked EPSCs in spinal cord slices of healthy and carrageenan-

given rats. The role of CB1 and TRPV1 receptors in the action of 20:4-NAPE has been investigated

by using the respective selective antagonists. This electrophysiological study is sounding

interesting in discovering the multiple target-induced action of endocannabinoids , however

criticisms are below reported.

20:4-NAPE is supposed to be the precursor of anandamide in all the enzymathic pathway through

which anandamide is synthesized. The authors mention the review by Lu and Mackie (2016), which

illustrates multiple enzymathic pathway without any specific mention of 20:4-NAPE. Since this is

crucial for the study meaning the enzymathic pathway from 20:4 NAPE to AEA and its occurrence in

in vitro preparations should be better described. Is there any evidence showing that NAPE PLD-

inhibitors antagonize the action of 20:4-NAPE? The authors also state that: “…..a direct effect of

20:4-NAPE cannot categorically excluded…”….concerning this…. is there any evidence of 20:4 NAPE

binding to CB1 or TRPV1 receptors? A possible action on PPARa and PPARg receptors by

anandamide should be also taken in consideration!

We agree. References as suggested have been added.

We have added new evidence of increased anandamide production after 20:4-NAPE application under

our experimental conditions with acute spinal cord slices. In these experiments there was a clear

concentration dependence of the anandamide production. Also, local concentration in the slice will be

most likely much higher, compared to the bath solution that was analyzed. Varga and colleagues

(2014) have previously provided compelling evidence that 20:4-NAPE is converted to anandamide by

Page 3 of 76



For Peer Review

4

cultured primary sensory neurons. They also demonstrated that primary sensory neuron cultures

express multiple pathways of anandamide production. To the best of our knowledge no inhibitors for

any enzymes potentially involved in anandamide synthesis, but SHIP-1, are available. Varga and

colleagues tried to study the contribution of the various pathways by down-regulating enzyme

expression (personal communication). However, while neurons have been transfected successfully at

least with the scrambled siRNA, no significant reduction in any of the enzymes was achieved within a

week.

We have no evidence available at this moment that would suggest or contradict direct effect of 20:4-

NAPE on the TRPV1 and CB1 receptors. Only circumstantial evidence is available on the absence of

direct effect of 20:4-NAPE on the TRPV1 receptor. 20:4-NAPE did not activate TRPV1 receptors in DRG

cultures at room temperature, in contrast to responses evoked by capsaicin application (Varga et al.

2014). Hence 20:4-NAPE is unlikely to produce any effect until it is converted to anandamide in a

concentration enough to induce biological effects. However, as no data on comprehensive screening

for the binding/effect of 20:4-NAPE is available, an action by 20:4-NAPE on some receptors/ion

channels cannot be categorically excluded.

We agree that possible effects through PPAR receptors should be also considered. However, PPAR are

nuclear receptors and slower effect of its activation could be expected. In our study we observed

rather immediate effect on synaptic transmission after 20:4-NAPE application. Another study will be

needed to look into the possible interaction with the PPAR receptors.

Another issue is related to the concentratiuons of 20:4-NAPE, CB1 and TRPV1 antagonists. Is there

any evidence of conc.-dependent effect of 20:4 NAPE? How have the concentration of 20:4-NAPE,

CB1 and TRPV1 antagonists been chosen?

We agree. We have added information into the methods section concerning the antagonists used.

The concentration of NAPE used was based on data from experiments on DRG cultures and also from

our preliminary data. Our new data on anandamide production also show concentration dependent

effect and support the 20:4-NAPE concentration used in our experiments. Varga and colleagues have

previously reported a concentration-dependent effect of NAPE application on responses in primary

sensory neurons. Similarly to the greater anandamide release with concentrations higher than 20 uM,

we found in our preliminary study that the inhibitory effects on EPSC is also higher with 200 uM than

20 uM 20:4NAPE. However, as 20:4-NAPE at higher concentrations may possibly produce more off

target effect, we decided to use a concentration which already produced significant changes without

apparent off target effects. With the long experimental protocol used, while recording from visually

identified neurons in spinal cord slices, it would prove very difficult to do any concentration

experiments with all the pharmacological treatments.

How have drugs been solved?

This is described in the Methods / materials section.

Do higher CB1 and TRPV1 conc. affect ongoing and evoked EPSCs frequency/amplitudes in normal

or inflammatory conditions? References need about PF514273 and SB366791 selectivity.

The references to the selectivity of the antagonists used were added to the methods section. We have

aimed / used high enough concentration to block all the activity of the given receptors.

The activity of TRPV1 channel in inflammatory conditions only deserves further investigation by

western blot analysis. Is TRPV1 expression changed in inflammatory vs healthy conditions? Is any

Page 4 of 76



For Peer Review

5

evidence about carrageenan pain models? Changes of TRPV1 expression in supraspinal areas

affecting pain have been reported in chronic pain model (i.e. Giordano et al.,2012).

We agree. This issue has been investigated earlier. We have added more references that demonstrate

increased TRPV1 expression in the DRG and spinal dorsal horn during inflammatory conditions

(Amaya et al 2003, Luo et al. 2004, Kwon et al. 2014). Recently, up-regulation of TRPV1 receptors was

demonstrated using western blot analysis and immunohistochemistry in DRG also in carrageenan

inflammatory model (Kwon et al. 2014).

Which sections of the spinal cord have been slices cut from (L4-L6)?

Lumbar enlargement segments L4-L5 were used. It is now specified in the methods.

Abstract:

“The transient receptor potential vanilloid receptor “: The transient receptor potential vanilloid

type 1”

We agree. It was changed to the official name: transient receptor potential cation channel, subfamily

V, member 1

Reviewer: 2

The manuscript by Nerandzic et al. examines the effects of anandamide signaling on synaptic input

onto superficial lamina projection neurons that are presumably nociceptive 2nd order afferents.

The authors are actually applying a precursor of anandamide, 20:4-NAPE that they presume is

being converted into anandamide by the spinal cord cells. Application of 20:4-NAPE depresses

evoked EPSCs and reduces the frequency of spontaneous EPSCs in both naïve animals and animals

that have undergone a peripheral inflammatory sensitization protocol. However, while CB1-Rs

mediate this depression in synapses from naïve animals, both CB1-Rs and TRPV1 appear to

contribute to this depression in inflammation sensitization animals. This is a really interesting (and

well-written) study. The observation that inflammation-induced sensitization changes the

contribution of TRPV1 channels to endocannabinoid-mediated modulation is very interesting.

However, there are a number of issues that need to be resolved before the manuscript can be

effectively evaluated.

Major concerns

1. Statistical Issues – First, the authors report the number of cells recorded from, but do

not report the number of animals. That appears to be a requirement of the based on

the BJP statement at the end of this submission. Second, the authors report the use of

1- & 2-way ANOVAs, t-tests and non-parametric rank tests, but never identify what

tests are used in which experiment/figure. Furthermore, all that is presented from the

statistical test are the p-values. The authors need to report the test statistic (e.g., the F

or t values) along with the degrees of freedom. There are also a number of cases where

it isn’t clear what groups are being compared statistically. An example is on pg. 9, 1st

paragraph where there is an increase in sEPSC frequency in the PF-treated group that

was not statistically significant, but I don’t see a vehicle control.

We agree. We have added the number of animals in the methods/animals section. Details of the

statistical methods used were added to the figure legends.

Page 5 of 76



For Peer Review

6

On page 9, 1st

paragraph the control for the PF treatment (100%) was the frequency of sEPSCs

recorded before the PF application in the mentioned case. The experimental protocol used during

recording from one neuron consisted: control (recording solution, 4min) - PF514273 application (in

recording solution, 6min) - PF514273 + 20:4-NAPE application (in recording solution, 4min) –

capsaicin application (in recording solution, 3min). Recording solution contained bicuculline and

strychnine during the whole recording to block inhibitory synaptic transmission. Repeated measure

ANOVA on ranks was used to compare the three different experimental conditions (ctrl, PF514273,

PF514273+20:4-NAPE) applied subsequently.

2. What exactly are the spontaneous EPSCs? They are spontaneous and very small

compared the eEPSCs. Are they mini EPSCs? If so, were they recorded in TTX? There is a

previous article by one of the co-authors (Spicarova D, Palecek J (2009) where they

made similar recordings that were in TTX. I usually think of sEPSC recordings as being

made in normal ACSF that are due to spontaneous activity (action potentials) of various

presynaptic neurons. If this is what is being recorded, would the effects of 20:4-NAPE

be due to changes in excitability in these presynaptic neurons?

The spontaneous EPSCs recorded in our experiments were spontaneous excitatory currents recorded

in the superficial (lamina I. II.outer) dorsal horn neurons in normal ACSF. We did not record miniature

currents with the TTX added in the experiments described here. The activity comes from presynaptic

endings of primary afferents, descending pathways and dorsal horn neurons. All the inhibitory

transmission was blocked by the presence of bicuculline and strychnine. The changes in the EPSC

frequency recorded are due to changes in the presynaptic endings. The evoked currents eEPSCs, were

evoked by dorsal root stimulation, activating primary afferents. There is documented effect of applied

anandamide on the sodium channels (Kim et al., 2005) and T-type Ca2+ channels (Chemin et al.,

2001) that regulate neuronal excitability. We cannot exclude the possibility that there was some

effect on these channels in our experiments beside the effect on TRPV1 and CB1 receptors and we

have added that information in the discussion.

3. I compliment the authors for reporting the number of neurons that did and did not

respond to the treatment, but there is something about the way they report this that

does not quite add up. For example, on pg. 13, 1st paragraph the authors report that

the eEPSC amplitude decreased to 78.5+/- 6.6% in 14 neurons. They then say that a

decrease of >15% was observed in 9 out of the 14 animals. This would suggest to me

that there is quite a lot of variability in these recordings if one is to get an average of

78% with some synapses decreasing much more than 15% and others decreasing less

than 15% as the authors stated. The problem is that the standard error seems to be too

small to get the observed level of depression. Am I missing something? Can the authors

re-check these stats?

We did not found any errors in our data. Please find below the data and the statistical analysis for the

14 neurons mentioned in your comment.

Ctrl 20:4-NAPE

100.00 92.58

100.00 99.36

100.00 104.81

100.00 63.30

100.00 119.60

100.00 63.21

100.00 48.16

Page 6 of 76



For Peer Review

7

100.00 83.03

100.00 101.36

100.00 84.64

100.00 75.84

100.00 68.58

100.00 69.60

100.00 25.57

Descriptive Statistics:

Column Size Missing Mean Std Dev Std. Error C.I. of Mean

20:4-NAPE 14 0 78.546 24.757 6.616 14.294

Column Range Max Min Median 25% 75%

20:4-NAPE 94.035 119.603 25.569 79.438 63.296 99.361

Column Skewness Kurtosis K-S Dist. K-S Prob. Sum Sum of Squares

20:4-NAPE -0.440 0.287 0.125 0.703 1099.639 94339.417

4. The authors have an important observation in that TRPV1 appears to have opposing

effects on synaptic transmission. They report that the TRPV1 inhibitor, SB366791,

increases sEPCS frequency, suggesting a tonic TRPV1 that enhances synapses. However,

SB also blocks the synaptic depression elicited by 20:4-NAPE. These opposing effects of

TRPV1 need to be more fully address. For example, are the sEPCS coming from a

different source than the eEPSCs? Perhaps excitatory interneurons that are part of

polysynaptic inputs to these projection neurons? Another possibility is that TRPV1 is

depressing inhibitory interneurons and therefore the increase in sEPSCs is a

disinhibitory effect. Kim et al. (Neuron 74: 640) has reported that TRPV1 channels are

present on inhibitory interneurons and mediate synaptic disinhibition that contributes

to allodynia.

In our experiments TRPV1 antagonist SB366791 had inhibitory effect on sEPSC frequency after

inflammation (Fig 5D). This is likely due to tonic activity of presynaptic TRPV1 receptors after

inflammation induction. The NAPE application following the SB treatment does not lead to a

significant decrease of the sEPSC frequency. The effect on the inhibitory interneurons should be

prevented by the presence of bicuculine and strychnine in the solution, so even if there would be

effect of TRPV1 activation on the inhibitory neurons, it should not be evident in the recordings. It was

shown before that activation of presynaptic TRPV1 receptors by capsaicin may lead to depression of

evoked excitatory postsynaptic currents in the spinal cord dorsal horn while spontaneous activity was

increased (Baccei et al 2003). It is thus possible that similar mechanism was involved in our

experiments. While it would be of our great interest to clarify all the mechanisms involved in this, we

feel that it would need another set of new experiments to possibly answer that question. The evoked

currents were present after activation of primary afferents in the dorsal root. The population of the

currents present during recording of the spontaneous activity could have other sources including

dorsal horn interneurons and endings of the descending pathways. Based on the evidence available,

we assume that presynaptic TRPV1 receptors are present only on the primary afferents endings.

Another factor in our experiments could be the plasticity of the anandamide synthesizing pathways.

Measurements were made 24 hours after the induction of inflammation. We know that

transcriptional changes are associated with the inflammation and there are changes in NAPE-PLD

expression in the DRG neurons (Suosa-Valente et al., 2017). Changes in the local production of

Page 7 of 76



For Peer Review

8

anandamide after NAPE application, could affect the changes seen in the control and inflammatory

conditions in our experiments.

I was happy to see the authors cite Yang et al. (1999) which show that TRPV1 activation depresses

synaptic transmission. The authors should take a closer look at the Baccei et al. (2003) cite which I

think also shows that TRPV1 activation increases frequency of minis, but also depresses eEPSC

amplitude (Fig. 5A). I would also like to point the authors attention to an article by the Michael

Andresen lab (J Neurosci 36:8957) which examines Ca2+ “nanodomains” and appears to show that

Ca2+ entry TRPV1 does not impact synchronous neurotransmitter release. This may explain why

TRPV1 activation can have one effect on spontaneous release and an alternative effect on evoked

release.

We agree. We are well aware of the paper by Baccei et al 2003 that was already cited here. The role

of TRPV1 receptors on presynaptic endings in modulation of neurotransmitter and neuromodulators

release is of high interest to us, and seems to be quite complicated with not enough evidence. We

thank the reviewer for pointing out the new interesting paper about the calcium nanodomains

involved in transmitter release under different modes of synaptic transmission and the potential role

of TRPV1 receptors in this process. It clearly demonstrates the complexity of neurotransmitter release

under different conditions and may help us to explain some of our other results. However, we feel

that it is out of scope for discussion in this paper.

Minor Concerns

1. Pg. 4, 2nd paragraph: Is this true? Aren't there TRPV1 channels on the free nerve

endings in the periphery?

We agree that the sentence was not well phrased. We have corrected this imprecision induced by

word count text reduction.

2. Same paragraph: Just a suggestion, but perhaps the authors want to include citations

that document synaptic depression as a result of TRPV activation.

We cite these papers in the discussion. Unfortunately it is difficult to add these to the introduction due

to the word count limitations.

3. Pg. 6, 1st paragraph: “intensity ranging between 20 and 350 μA were applied” This

would seem to be a very broad range for stimulating the dorsal root nerve (is there

that much variability in the suction electrode seal around the nerve?). Is it possible that

the authors recorded EPSCs elicited by A-beta fibers when they used the lower end of

this range of stimuli? Is there any relationship between the heterogeneity in how

synapses responded to the various treatments and the stimuli used to get the evoked

transmission? Perhaps synapses that were not depressed or even potentiated by 20:4-

NAPE were A-beta, polysynaptic inputs and not C fiber inputs.

The variability of the stimulating current is due to many factors including the profile of the suction

electrode, the size and condition of the dorsal root etc. We report the whole range of the stimulating

currents used, while most of the recordings were made in the higher range. We have used 0.5ms long

stimulus that in our experience leads to activation of slow conduction velocity afferents, compared to

the 0.1ms stimulation pulse. Also the neurons in the superficial laminae tend to receive majority of

their input from nociceptive primary afferents. We did not find any correlation between the current

used and the effect of NAPE. However, we are aware of this and we cannot completely rule out that

some of the recorded currents had component of low threshold afferents that do not express TRPV1

receptors.

Page 8 of 76



For Peer Review

9

4. I think that the authors might benefit from explaining their rationale for applying 20:4-

NAPE in the Introduction (or the beginning of the Results) as opposed to the Discussion

section. When I was reading the manuscript, I kept asking myself “why didn’t they just

apply anandamide?”. However, when I read the rationale for this approach in the

Discussion section, I thought it was a valid approach and wished that I had known the

authors’ reasoning earlier.

We agree. We have added some of the reasoning in the abstract and in the introduction.

5. Pg.16, 2nd paragraph; The authors state that myelinated afferents could contribute to

the eEPSC and that these do not express TRPV1. This needs a citation. Also, are the

authors referring to A-delta or A-beta fibers? If these are A-beta fibers, then the

contribution to the projection neurons should be polysynaptic. Is it possible that the

excitatory polysynaptic interneurons may express TRPV1 channels? As noted earlier,

from the Kim et al 2012 reference, inhibitory interneurons do express TRPV1.

We have added citation on the TRPV1 receptors expression in DRG neurons. As was discussed above,

with electrical stimulation we cannot exclude the possibility of activating TRPV1 negative fibres. As

was also discussed, the possible effect on inhibitory interneurons possibly expressing TRPV1 receptors

(there were no postsynaptic TRPV1 receptors in the spinal cord detected in the study by Cavanaugh et

al 2011) should be highly limited in our study, due to presence of bicuculline and strychnine in our

bath solution.

6. Pg.16, 3rd paragraph; What is meant by “"increasing the action potential firing of the

postsynaptic cell is preserved"? Did the authors mean the presynaptic cell?

The text was now shortened, as this explanation of the Baccei experiment was not critical for this

paper and the word counts prevented us to explain it in detail.

Reviewer: 3

This study aims to study the effect of 20:4-NAPE application on the excitatory synaptic

transmission in the dorsal horn of the spinal cord.

Despite experiments of very good quality, author’s conclusions are not fully supported by results.

Authors shown that 20:4-NAPE decreased sEPSCs frequency and eEPSCs amplitude and conclude

that 20:4-NAPE is use as a precursor for synthesis of anandamide that acts on presynaptic CB1 and

or TRPV1 receptors.

Authors suggested that observed effects are due to endogenous synthesis of anandamide from

20:4-NAPE. But there is no evidence so far of such a synthesis in this slice model. A

pharmacological study of this synthesis would be needed for a clear demonstration. Also authors

do not discuss a potential direct effect of 20:4-NAPE.

We agree. We have now added new evidence of anandamide production under our experimental

conditions with acute spinal cord slices. The results clearly show 20:4-NAPE concentration dependent

anandamide production. There is also compelling evidence from Varga and co-workers that the

application of 20:4-NAPE to DRG cultures results in anandamide production.

Page 9 of 76



For Peer Review

10

Also, accepting endogenous biosynthesis of anandamide from 20:4-NAPE, effects of CB1 activation

on the excitatory synaptic transmission have already be shown.

We agree that the effect of anandamide on CB1 activation in spinal cord slice preparations has been

investigated previously. However, so far the effect of exogenous anandamide has been studied. Here

we studied, for the first time, the effect of endogenous anandamide synthesized from exogenous

precursor on that synaptic transmission. Here, anandamide is synthesized only by those neurons

which have the appropriate enzymatic machinery. Also these enzymes may change their properties

and expression during inflammation. As a lipophilic agent, anandamide is expected to act close to its

source, as opposed to flooding the preparation with exogenous anandamide. Hence, we feel that the

findings we report here are novel.

So the novelty of this study is the involvement of both CB1 and TRPV1 receptors under

inflammatory condition. But TRPV1 mediated mechanism explaining these results is not

demonstrated. Authors suggested that under inflammatory conditions anandamide synthesis from

20:4-NAPE would be increased nearby TRPV1 on primary afferent fibers terminals. However CB1

receptors involved in the decreased of EPSCs are also expressed on primary afferent fibers

terminals. Authors should discuss this point.

As we wrote above, we would respectfully argue for the 20:4-NAPE application induced inhibitory

effect is being a novel finding as well. Regarding the possible mechanisms, here we can only

speculate, because we have too few information on anandamide synthesis in the dorsal horn of the

spinal cord. However, we have added new experiments that demonstrate anandamide synthesis after

20:4-NAPE application under our experimental conditions. Based on PCR and immunostaining

experiments, it is highly likely that several cell types, including neurons, astrocytes and microglia

contribute to 20:4-NAPE-induced anandamide production through multiple pathways. However, the

spatial distribution of the enzymes and targets are not known, though this information is crucial as

anandamide is expected to act close to its source. There is increased expression of TRPV1 and their

phosphorylation after inflammation and this may change the interaction between the anandamide

produced from 20:4-NAPE in their favour compared to the CB1.

Authors did not specify the nature of eEPSC recorded. It is not clear if authors recorded specifically

neurons directly postsynaptic to afferent fibers by selecting neurons receiving C-fibers or Ad-fibers

monosynaptic eEPSC. These information would help to better understand results of this study.

We have recorded from neurons in the dorsal horn superficial laminae (I and outer II). These neurons

are post-synaptic to primary afferent terminals, mostly A-delta and C-fibres. The stimulation

parameters were set to activate C-fibres. Based on our previous published experiments and also

based on capsaicin application in the end of experiments in this paper, almost all of the neurons in

this area receive capsaicin sensitive input. This would suggest that most of the recorded evoked

currents were from nociceptive primary afferents. As we stated in the paper, we cannot exclude

completely activation of other afferent fibres.

Also miniature EPSC analysis would have been a good argument to confirm the presynaptic effect

of 20:4-NAPE application.

We agree. We have preliminary evidence with mEPSC recorded in the presence of TTX that shows

similar inhibitory effects of 20:4-NAPE application under our experimental conditions. However, we

feel that these these preliminary data require more experiments and a separate publication.

Page 10 of 76



For Peer Review

11

The overall physiological effect of endocannabinoid in the transmission and the modulation of

nociceptive information in the dorsal horn in naïve Vs inflammatory conditions should be discuss

including papers that have shown a role of spinal endocannabinoids in the spinal sensitization.

We agree. However, while this issue is of high importance, the complexity of endocannabinoid

signaling in spinal nociceptive processing is far from being elucidated. There are a large number of

unanswered questions in this context: for example, we do not know details (molecular, cellular) of

endocannabinoid production/hydrolysis/uptake, inflammation-induced plasticity of those, regulation

of endocannabinoid synthesis, or spatial relationship of various components of the endocannabinoid

endovanilloid systems. Without those details, the question how the spinal endocannabinoid system

can be utilized to reduce pain cannot really be answered. Our present work is a step into that

direction.

While the CB1 receptor is believed to be an inhibitory receptor, a series of evidence indicate that it

may increase the activity of TRPV1. In agreement with the inhibitory effect, CB1 receptor activation

including that by anandamide reduces transmitter release from spinal terminals of primary afferents

(Morisset and Urban, 2001). Consistently, exogenous anandamide, has been shown to reduce,

through the CB1 receptor, nociceptive stimulation-evoked responses in the spinal cord of rats

following intraplantar carrageenan injection (Harris et al., 2000). Inhibiting FAAH activity, believed to

act through increasing spinal anandamide levels, also reduces inflammatory pain-related behaviour

following intraplantar carrageenan injection (Okine et al., 2012). However, this effect is produced by

a single but not by repeated injection of the FAAH inhibitor (Okine et al., 2012). While deleting FAAH

is associated with increased anandamide levels and an antinociceptive phenotype, FAAH-KO mice

exhibit greater responses to capsaicin that can be reduced by inhibiting CB1 TRPV1 (Carey et al.,

2016). Although, anandamide levels are increased 18 hours after intraplantar carrageenan injection

in the spinal cord (Buczynski et al., 2010) the source of anandamide and effect of that anandamide is

not known. In addition, CB1 receptor activation on inhibitory interneurons facilitates excitatory

responses in the spinal dorsal horn. This already complex picture is just further complicated by the

responsiveness of TRPV1 to anandamide, and the expression of TRPV1, in addition to primary sensory

neuron terminals, by a group of inhibitory interneurons, where TRPV1 exerts an inhibitory effect.

Further, other anandamide-responding receptors are also expressed in the spinal dorsal horn, and in

addition to anandamide, other endovanilloids are also synthesized.

Minor points:

- Frequency values should be indicated in the text

We have added frequency in Hz of the naïve and the inflammatory groups of neurons.

- Amplitudes values are indicated in the text with positive numbers whereas EPSCs are

inwards currents. This should be corrected

As we agree with the reviewer that the recorded currents are inward, we do not see much relevance

in giving all the amplitudes with minus sign. Also in the published literature the amplitude of the

currents is usually given as a positive value.

- Some writing need to be corrected. For example in introduction, third paragraph: authors

probably mean “CB1 receptor expression has been detected in various structures”

Page 11 of 76



For Peer Review

12

We agree. We have attempted to rectify all those issues.

Page 12 of 76



For Peer Review

1

Peripheral inflammation alters N-arachidonoylphosphatidylethanolamine (20:4-NAPE)

induced modulation of nociceptive spinal cord synaptic transmission

20:4-NAPE-modulated spinal synaptic transmission

Vladimir Nerandzic1, Petra Mrozkova

1, Pavel Adamek

1.3, Diana Spicarova

1,

Istvan Nagy2 and Jiri Palecek

1

1Department of Functional Morphology, Institute of Physiology,

The Czech Academy of Sciences, Prague, Czech Republic

2Department of Anaesthetics, Pain Medicine and Intensive Care, Imperial College London,

Faculty of Medicine, Chelsea and Westminster Hospital, London, UK

3Faculty of Science, Charles University, Prague, Czech Republic

Corresponding author:

Jiri Palecek M.D., Ph.D.

Department of Functional Morphology

Institute of Physiology Czech Academy of Sciences

Videnska 1083

142 20 Praha 4

Czech Republic

e-mail: [email protected]

office: (-420) 2 4106 2664

fax: (-420) 2 4106 2488

Page 13 of 76



For Peer Review

2

Abstract

Background and Purpose

Endocannabinoids play an important role in modulating spinal nociceptive signalling, crucial

for the development of pain. The cannabinoid receptor 1 (CB1) and the transient receptor

potential cation channel subfamily V member 1 (TRPV1) are both activated by the

endocannabinoid anandamide that is a product of biosynthesis from the endogenous lipid

precursor N-arachidonoylphosphatidylethanolamine (20:4-NAPE). Here we are first to report

CB1 receptor- and TRPV1-mediated effects of 20:4-NAPE application on spinal synaptic

transmission in control and inflammatory conditions.

Experimental Approach

Spontaneous (sEPSCs) and dorsal root stimulation-evoked (eEPSCs) excitatory postsynaptic

currents from superficial dorsal horn neurons in rat spinal cord slices were assessed.

Peripheral inflammation was induced by carrageenan. Anandamide concentration was

assessed by mass spectrometry.

Key Results

Application of 20:4-NAPE increased anandamide concentration in vitro. 20:4-NAPE (20 µM)

decreased sEPSCs frequency and eEPSCs amplitude in control and inflammatory conditions.

The inhibitory effect of 20:4-NAPE was sensitive to CB1 antagonist PF514273 (0.2 µM) in

both conditions, but to the TRPV1 antagonist SB366791 (10 µM) only after inflammation.

After inflammation 20:4-NAPE increased sEPSCs frequency in the presence of PF514273 and

this increase was blocked by SB366791.

Conclusions and Implications

While 20:4-NAPE treatment produced an inhibitory effect on excitatory synaptic transmission

in both naive and inflammatory conditions, peripheral inflammation altered the underlying

mechanisms. Our data indicate that 20:4-NAPE application induced mainly CB1 receptor-

mediated inhibitory effects in naive animals while TRPV1-mediated mechanisms were also

involved after inflammation. Increasing anandamide levels for analgesic purposes by applying

substrate for its local synthesis may be superior to systemic anandamide application or

inhibition of its degradation.

Page 14 of 76



For Peer Review

3

Abbreviations

20:4-NAPE, N-arachidonoylphosphatidylethanolamine

AEA, anandamide, N-arachidonoylethanolamine

CB1, cannabinoid receptor 1

DRG, dorsal root ganglion

NAPE-PLD, N-acyl phosphatidylethanolamine phospholipase D

PWL, paw withdrawal latency

TRPV1, transient receptor potential cation channel subfamily V member 1

Tables of Links – instructions within the submission process

TARGETS

GPCRs

CB1 – ID 56

Voltage-gated ion channels

TRPV1 – ID 507

Enzymes

NAPE-PLD

(N-Acylphosphatidylethanolamine-phospholipase D)

LIGANDS

Anandamide – ID 2364

20:4-NAPE

PF514273, 2-(2-chlorophenyl)-3-(4-chlorophenyl)-7-(2,2-difluoropropyl)-6,7-dihydro-2H-

pyrazolo[3,4-f][1,4]oxazepin-8(5H)-one

SB366791 – ID 4309, (2E)-3-(4-chlorophenyl)-N-(3-methoxyphenyl)prop-2-enamide

Page 15 of 76



For Peer Review

4

Introduction

Modulation of synaptic transmission in the spinal dorsal horn is pivotal in nociceptive

signalling. An important part in this process is mediated by the transient receptor potential

cation channel, subfamily V, member 1 (TRPV1) and the Gi/o protein-coupled cannabinoid

type 1 (CB1) receptors (Katona and Freund, 2008; Nagy et al., 2014; Sousa-Valente et al.,

2014a). However, our current understanding does not allow us to utilise the putative analgesic

potential offered by controlling spinal signalling through these two receptors, fully.

Spinal TRPV1 is expressed predominantly by the central branches of nociceptive small- and

medium-sized dorsal root ganglion (DRG) neurons (Caterina et al., 1997). TRPV1 activation

has an excitatory effect through increasing the transmitter release from the terminals of DRG

neurons, but may depress evoked currents (Baccei et al. 2003).

CB1 receptor expression has been detected in various structures including inhibitory neurons,

astrocytes and central terminals of nociceptive small to medium size DRG neurons

(Ahluwalia et al., 2000; Alkaitis et al., 2010; Veress et al., 2013). Activation of both

presynaptic CB1 receptors and CB1 receptors on inhibitory interneurons leads to reduced

transmitter release from the respective neurons (Morisset and Urban, 2001; Pernia-Andrade et

al., 2009).

Several endogenous agents including N-arachidonoylethanolamine (anandamide; AEA)

activate both TRPV1 and the CB1 receptors (Ahluwalia et al., 2003; Devane et al., 1992;

Tognetto et al., 2001; Zygmunt et al., 1999). Importantly, sub-populations of DRG neurons as

well as spinal cord cells are able to synthesise or degrade anandamide (Carrier et al., 2004;

van der Stelt et al., 2005; Varga et al., 2014; Vellani et al., 2008).

Anandamide synthesis occurs through multiple metabolic pathways either in a Ca2+

-

insensitive or Ca2+

-sensitive manner (Ueda et al., 2013). N-

arachidonoylphosphatidylethanolamine (20:4-NAPE) constitutes the precursor for

anandamide synthesis in all pathways (Snider et al., 2010; Ueda et al., 2013; Wang and Ueda,

2009). We have shown recently that application of 20:4-NAPE to cultured DRG neurons

results in anandamide production in a concentration and temperature-dependent manner

(Varga et al., 2014).

Page 16 of 76



For Peer Review

5

Painful peripheral pathologies including tissue inflammation alter the expression and/or

activity of both TRPV1 and the CB1 receptor (Amaya et al., 2003; Amaya et al., 2006; Kanai

et al., 2005; Kwon et al., 2014; Luo et al., 2004; Richardson et al., 1998; Spicarova et al.,

2011; Spicarova and Palecek, 2009). Inflammation also induces changes in anandamide levels

in the spinal cord (Buczynski et al., 2010; Costa et al., 2010) and the expression of

anandamide synthesising and hydrolysing enzymes in DRG neurons (Lever et al., 2009;

Sousa-Valente et al., 2017). Here, for the first time, instead of “flooding” the entire

preparation by exogenous anandamide, we studied how providing substrate (20:4-NAPE) for

anandamide-synthesising pathways in the spinal cord affects nociceptive spinal synaptic

transmission and what role TRPV1 and CB1 receptors play in that process under naive and

inflammatory conditions.

Methods

Animals

All animal care and experimental procedures were in accordance with local Institutional

Animal Care and Use Committee and consistent with the guidelines of the International

Association for the Study of Pain, the U.K. Animals (Scientific Procedures) Act (1986) and

EU Directive 2010/63/EU for animal experiments. Animal studies are reported in compliance

with the ARRIVE guidelines (Kilkenny et al., 2010; McGrath and Lilley, 2015). Altogether

49 male Wistar rats (Institute of Physiology CAS, Czech Republic) of postnatal days (P19-

P23) were used in this study. Animals were maintained under temperature (22±2°C) and light-

controlled (12 h light/dark cycle) conditions with free access to food and water.

Spinal cord slice preparation

Male Wistar rats (P19-P23) were anaesthetised with isofurane (3%), the lumbar spinal cords

were removed and immersed in oxygenated ice-cold dissection solution containing (in mM)

95 NaCl, 1.8 KCl, 7 MgSO4, 0.5 CaCl2, 1.2 KH2PO4, 26 NaHCO3, 25 D-glucose and 50

sucrose. Animals were sacrificed by subsequent medulla interruption and exsanguination.

Each spinal cord was fixed to a vibratome stage (VT 1000S, Leica, Germany) using

cyanoacrylate glue in a groove between two agar blocks. Acute transverse slices 300-350 µm

thick were cut from L4-L5 segments, incubated in the dissection solution for 30min at 33°C,

stored in a recording solution at room temperature and allowed to recover for 1 h before the

electrophysiological experiments. Recording solution contained (in mM) 127 NaCl, 1.8 KCl,

1.2 KH2PO4, 2.4 CaCl2, 1.3 MgSO4, 26 NaHCO3, and 25 D-glucose. For the actual

Page 17 of 76



For Peer Review

6

measurement, slices were transferred into a recording chamber that was perfused continuously

with recording solution (room temperature) at a rate of ~2 ml.min-1

. All extracellular solutions

were saturated with carbogen (95% O2, 5% CO2) during the whole process.

Patch-clamp recording

Altogether, sEPSCs were recorded from 98 and eEPSCs from 79 superficial dorsal horn

neurons. Individual neurons were visualized using a differential interference contrast

microscope (DM LFSA, Leica, Germany) equipped with a 63×0.90 water-immersion

objective and an infrared-sensitive camera (KP-200P, Hitachi, Japan) with a standard

TV/video monitor (Hitachi VM-172, Japan). Patch pipettes were pulled from borosilicate

glass tubing when filled with intracellular solution; they had resistances of 3.5-6.0 MΩ. The

intracellular pipette solution contained (in mM) 125 gluconic acid lactone, 15 CsCl, 10

EGTA, 10 HEPES, 1 CaCl2, 2 MgATP, and 0.5 NaGTP and was adjusted to pH 7.2 with

CsOH. Voltage-clamp recordings in the whole cell configuration were performed with an

AxoPatch 200B amplifier (Molecular Devices, Sunnyvale, CA, USA) at room temperature

(21-24°C). Whole cell responses were low-pass filtered at 2 kHz and digitally sampled at 10

kHz. The series resistance of the recorded neurons was routinely compensated by 80% and

was monitored during the whole experiment. AMPA receptor-mediated spontaneous

excitatory postsynaptic currents (sEPSCs) were recorded from superficial dorsal horn neurons

in laminae I and II(outer), clamped at -70 mV in the presence of 10 µM bicuculline and 5 µM

strychnine. To evoke EPSCs, a dorsal root was stimulated with a suction electrode using a

constant current isolated stimulator (Digitimer, DS3, Hertfordshire, UK). Test pulses of

0.5 ms duration and an intensity ranging between 20 and 350 µA were applied at a frequency

of 0.033 Hz. The intensity of the stimulation was adjusted to evoke stable EPSCs. Application

of each drug lasted for 4 minutes period (recording solution, 20:4-NAPE, capsaicin, co-

application 20:4-NAPE and PF514273, co-application 20:4-NAPE and SB366791) or 6

minutes antagonist pretreatment (PF514273, SB366791, co-application of PF514273 with

SB366791). Neurons with capsaicin sensitive primary afferent input were identified by

increase of sEPSC frequency (>20%) following capsaicin (0.2 µM) application at the end of

the experimental protocol. Capsaicin was applied in 87% of the recorded neurons and 92% of

these responded with sEPSC frequency increase. The software package pCLAMP version

10.0 (Axon Instruments, CA, USA) was used for data acquisition and subsequent off-line

analysis. Cells with a series resistance >20 MΩ were excluded from the analysis.

Page 18 of 76



For Peer Review

7

Anandamide release experiments

Spinal cord slices from 5 animals were prepared in the same way as for the

electrophysiological experiments. For each of the 5 experiments, 18 acute spinal cord slices

were used. Slices were put into plastic safe-lock tube with 200 µl of the recording solution,

saturated by carbogen (95% O2 - 5% CO2) during the whole process. Incubation of 10 min

was used before the whole volume of the solution (sample) was extracted and immediately

frozen for later mass spectrometry analysis. The solution in the tube was immediately

replaced with another solution sample. In each experiment 8 samples were taken after 10 min

of incubation: 1 (recording solution), 2 (recording solution), 3 (20:4-NAPE, 20 µM), 4

(recording solution), 5 (20:4-NAPE, 100 µM), 6 (recording solution), 7 (20:4-NAPE, 200

µM), 8 (recording solution). Additional samples with only 20:4-NAPE, 20-100-200 µM

without the slices were also prepared and analysed.

The collected samples were analysed for the presence of AEA with mass spectrometry. For

calibration purposes solutions of AEA and 20:4-NAPE were used. Anandamide content was

determined by reversed-phase high-performance chromatography using Agilent 1100 LC

system (Agilent, Palo Alto, CA, USA) consisting of a degasser, a binary pump, an

autosampler, and an thermostatic column compartment. Chromatographic separation was

carried out in a Kinetex 2.6u PFP, 100A column (100 x 2.1 mm I.D., Phenomenex, Torrence,

CA, USA). The sample (10 µl) was injected into the column and eluted with a gradient

consisting of (A) water-formic acid 100:0.1 v/v and (B) acetontrile-formic acid 100:0.085 v/v

(flow rate 0.35 ml.min-1

and temperature 40°C). The gradient started at A/B 80:20 for 5 min,

reaching 100% B after 10 min. For the next 5 min the elution was isocratic at 100% B.

Elution was monitored by an ion-trap mass spectrometer (Agilent LC-MSD Trap XCT-Ultra;

Agilent, Palo Alto, CA, USA). Atmospheric pressure ionization-electrospray ionization (API-

ESI) positive mode ion-trap mass spectrometry at MRM (multiple reaction monitoring) mode

was used with transition of m/z 348.1>287.1 for anandamide (retention time 11.2 min) when

monitored mass range was 100-400 m/z. Operating conditions: drying gas (N2), 12 l.min-1

;

drying gas temperature, 350°C; nebulizer pressure, 30 psi (207 kPa). The areas of the

anandamide peak were measured. The results from each experiment (peak areas of AEA)

were normalized to the AEA production after the 200 µM 20:4-NAPE application (100%).

Peripheral inflammation

Peripheral inflammation was induced in a group of animals 24 h before the spinal cord slice

preparation was made. Under isoflurane (3%) anaesthesia, both hind paws were injected

Page 19 of 76



For Peer Review

8

subcutaneously by a 3% mixture of carrageenan (50 µl) in a physiological saline solution. The

animals were left to recover in their home cages. This carrageenan injection in peripheral

tissue is thoroughly characterized and established animal model of inflammatory pain (Ren

and Dubner, 1999). Naive animals were used as controls.

Behavioural testing

The animals used in model of peripheral inflammation were tested for responsiveness to

thermal stimuli before and 24 h after the model induction. Paw withdrawal latencies (PWLs)

to radiant heat stimuli were determined for both hind paws using Plantar Test 37370 apparatus

(Ugo Basile, Italy). The rats were placed under non-binding, clear plastic cages on a 3 mm

thick glass plate and left to adapt at least for 20min. The radiant heat was applied to the

plantar surface of each hind paw until a deliberate escape movement of the paw was detected

by Plantar Test apparatus. The PWLs were tested 4 times for each hind paw with at least 5min

intervals between the trials. Results from each hind paw were averaged. Baseline withdrawal

latencies were determined in all animals before any experimental procedure.

Data and statistical analysis

The data and statistical analysis comply with the recommendations on experimental design

and analysis in pharmacology (Curtis et al., 2015). Some data were normalized as a

percentage of the control values (100%). Results are shown as means ± SEM. For offline

analysis of the recorded sEPSCs data, segments of 2-min duration were used for each

experimental condition. Only sEPSCs with amplitudes 5 pA or greater (which corresponded

to at least twice the noise level) were included in the frequency analysis. In the case of

amplitude analysis, the same sEPSCs events were used. Statistics were calculated using

SigmaStat 3.5 (Systat Software, CA, USA). A Kolmogorov-Smirnov test was used to evaluate

statistical significance for cumulative data. One-way ANOVA or one-way ANOVA repeated

measures (RM) followed by a post hoc test (Student-Newman-Keuls) or paired t-test was used

for data with normal distribution and non-parametric rank test or RM on ranks was used

where appropriate for statistical comparisons. P-value <0.05 was considered statistically

significant. Detailed information is given in the figure legends.

Materials

All chemicals used for extracellular and intracellular solutions were of analytical grade and

purchased from Sigma Aldrich (St. Louis, MO, USA) and Tocris Bioscience (Bristol, UK).

Page 20 of 76



For Peer Review

9

Capsaicin, SB366791 and PF514273 (Tocris Bioscience) and anandamide (Avanti Polar

Lipids, Alabaster, AL, USA) were dissolved in dimethylsulfoxide (DMSO; Sigma Aldrich),

which had a concentration <0.1% in the final solution. 20:4-NAPE (Avanti Polar Lipids) was

dissolved in chloroform, which had a concentration <0.1% in the final solution. Concentration

of selective TRPV1 antagonist SB366791 (10 µM) was based on our previous studies

(Spicarova et al., 2014a; Spicarova and Palecek, 2009) and the selectivity pA2 = 7.71

(Gunthorpe et al., 2004). Concentration of highly selective CB1 antagonist PF514273 (0.2

µM) was determined by considering Ki - 1 nM (Dow et al., 2009) and the needed diffusion

through the spinal cord slice. 20:4-NAPE was applied in the recording solution in

concentration (20 µM) based on our preliminary results and previous experiments performed

on DRG cultures (Varga et al., 2014). Carrageenan for induction of inflammation was

purchased from Sigma Aldrich.

Results

Application of 20:4-NAPE increased anandamide concentration in spinal cord slices

To verify the production of anandamide from 20:4-NAPE in our preparation, mass

spectrometry was used to analyse AEA content after application of different concentrations of

20:4-NAPE (20 µM, 100 µM and 200 µM) on spinal cord slices in vitro. Under the control

conditions with extracellular solution only, the average AEA concentration in the solution was

very low (7067±4532 of peak area, n=5, Fig. 1). AEA concentration increased gradually with

increasing concentration of 20:4-NAPE application (20 µM: 48324±27502; 100 µM:

103310±38179; 200 µM: 298004±139867 AEA peak area, n=5). To reduce the differences

between the individual experiments, the results were standardized for the statistical analysis

(Fig. 1). There was no AEA detected in the samples where 20:4-NAPE was present without

the slices. These results indicate that 20:4-NAPE (20 µM) application in our

electrophysiological experiments led to increased AEA concentration in the spinal cord slice.

Application of 20:4-NAPE reduced both spontaneous and evoked activity of spinal

dorsal horn neurons

The role of 20:4-NAPE in nociceptive synaptic transmission was investigated using

recordings of spontaneous EPSC and dorsal root stimulation-evoked EPSCs from neurons in

laminae I and II(outer) of the dorsal horn. Altogether the mean control frequency of sEPSCs

recorded in neurons from slices prepared from naive animals was 1.09±0.14 Hz (n=42).

Page 21 of 76



For Peer Review

10

Application of 20:4-NAPE (20 µM, 4min) robustly decreased sEPSC frequency in 12 of the

13 recorded neurons to 52.3±7.8% (n=13; Fig. 2A,B,C) when data was averaged from all

neurons in this group. The average amplitude of the sEPSCs was also reduced significantly

after the 20:4-NAPE application from 24.9±2.5 pA to 21.4±1.4 pA (n=13, p<0.05, paired t-

test). However, the decrease of the amplitude (>15%) was present in only 4 cells out of 13

neurons and the cumulative distribution of sEPSC amplitudes did not show significant change

after the 20:4-NAPE application (Fig. 2D).

Similarly to the changes in sEPSC’s, 20:4-NAPE (20 µM, 4min) also significantly decreased

the amplitude of eEPSCs to 70.5±9.0% (n=15; Fig. 2E,F). The reduction of eEPSCs

amplitude (>15%) was present in 8 of 15 recorded neurons. Together, these findings indicate

that application of 20:4-NAPE had a robust inhibitory effect on the excitation of superficial

spinal dorsal horn neurons in naive conditions.

The 20:4-NAPE-induced inhibitory effect on spontaneous activity was prevented by

blocking the CB1 but not TRPV1 receptors

In the next experiments we have investigated whether the 20:4-NAPE-induced inhibitory

effect is mediated through either of the anandamide’s main targets, the CB1 and TRPV1

receptors.

Application of the highly selective CB1 antagonist PF514273 (0.2 µM, 6min) caused a small

numerical increase of the sEPSCs frequency; however this enhancement did not reach

statistical significance (112.6±14.8%, n=11; Fig. 3A,C). Subsequent co-application of

PF514273 (0.2 µM) and 20:4-NAPE (20 µM, 4min) did not change the frequency of sEPSCs

compared to the control value and to the antagonist pretreatment (98.1±13.4%; Fig. 3A,C,F).

The amplitude of sEPSCs in control conditions (20.0±1.7 pA, n=11) was not affected either

by pre-application alone (18.8±1.8 pA) or the subsequent co-application of PF514273 and

20:4-NAPE (18.6±2.1 pA). The failure of 20:4-NAPE reducing either the frequency or the

amplitudes of sEPSCs in the presence of PF514273 is in contrast with the effect of 20:4-

NAPE alone (Fig. 2B).

Pretreatment with the selective TRPV1 antagonist SB366791 (10 µM, 6min) alone did not

change the sEPSC frequency (97.6±13.9%, n=10; Fig. 3B,D). However, co-application of

20:4-NAPE (20 µM, 4min) and SB366791 (10 µM) significantly reduced the frequency of

Page 22 of 76



For Peer Review

11

sEPSCs in 8 of 10 recorded neurons (53.0±14.7%, n=10; Fig. 3B, D) when compared to the

control values. The degree of reduction was not different from that 20:4-NAPE produced

alone (Fig. 2B). The amplitude of sEPSCs was not affected by the SB366791 pretreatment

(control: 29.0±4.1 pA, SB366791: 24.6±4.0 pA) or following SB366791 with 20:4-NAPE

treatment (25.2±4.6 pA, n=10).

Further, combined pretreatment with both CB1 and TRPV1 antagonists PF514273 (0.2 µM)

and SB366791 (10 µM, 6min) was tested. This pre-treatment caused a numerical increase of

the sEPSCs frequency that was not statistically significant (122.0±11.4%, n=8; Fig. 3E).

Subsequent co-application of both antagonists PF514273 (0.2 µM) and SB366791 (10 µM)

together with 20:4-NAPE (20 µM, 4min) did not change the frequency of sEPSCs compared

to the control value (100.2±7.6%; Fig. 3E). The average amplitude of sEPSCs was not

changed during entire experiment (control: 20.8±2.0 pA, PF514273 + SB366791: 19.2±1.6

pA, PF514273 + SB366791 + 20:4 NAPE: 17.4±1.1 pA, n=8).

To compare all the experimental situations and to diminish any influences of the antagonists’

applications alone, we have also analysed the data in a way where the antagonist application

together with 20:4-NAPE was expressed as a percentage of the previous condition (Fig. 3F).

PF514237 + 20:4-NAPE compared to the PF514273 pretreatment (87.5±5.3%, n=11);

SB366791 + 20:4-NAPE compared to the SB366791 pretreatment (58.1±16.0%, n=10) and

PF514237 + SB366791 + 20:4-NAPE compared to both antagonists pretreatment (84.6±6.6%,

n=8). This analysis confirmed the overall differential effect of the antagonists. These findings

taken together show that the 20:4-NAPE application-induced inhibitory effect on the

frequency of sEPSCs is mediated by activation of CB1 receptors, but not by TRPV1

receptors.

20:4-NAPE-induced reduction of eEPSCs amplitude was prevented by CB1 but not

TRPV1 antagonist in naive animals.

The respective antagonists of the CB1 and TRPV1 receptors PF514273 and SB366791 were

used to identify the contribution of these receptors to the 20:4-NAPE-induced decrease of

eEPSC amplitude. Application of PF514273 (0.2 µM, 6min) did not significantly change the

amplitude of eEPSCs (89.0±6.6%, n=13; Fig. 4A,C). Subsequent co-application of PF514273

(0.2 µM, 4min) and 20:4-NAPE (20 µM) did not change significantly the average amplitude

of the eEPSC compared to both the control (77.0±10.5%, n=13; Fig. 4C) and PF514273

Page 23 of 76



For Peer Review

12

pretreatment (88.5±10.1%; Fig. 4E) values. Out of these neurons 7 exhibited a lack of

reduction. Four of these 7 recorded neurons were not affected by 20:4-NAPE application and

in remaining 3 neurons the amplitude increased >15%. These results show that the inhibitory

effect of 20:4-NAPE on the eEPSCs amplitudes is mediated by CB1 receptors in the group of

superficial spinal dorsal horn neurons.

Inhibition of TRPV1 receptors by SB366791 (10 µM, 6min) alone did not change the eEPSC

amplitude (108.8±14.7%, n=10; Fig. 4B,D). Subsequent co-application of SB366791 (10 µM,

4min) and 20:4-NAPE (20 µM) did not prevent the amplitude decrease induced by the 20:4-

NAPE, compared to both control (64.9±9.7%, n=10; Fig. 4D) and SB366791 pretreatment

(64.2±9.3%; Fig. 4E) values. The amplitude reduction was evident in 8 of 10 neurons and it

did not change in the 2 remaining neurons. The degree of 20:4-NAPE-induced reduction in

the presence of SB366791 was not significantly different from that produced by 20:4-NAPE

alone (Fig. 4E). Hence, these findings indicate that TRPV1 is not involved in mediating the

20:4-NAPE-induced inhibitory effect on eEPSC amplitude.

Application of 20:4-NAPE reduced the frequency of sEPSCs in spinal dorsal horn

neurons under inflammatory conditions

Peripheral inflammation was induced by subcutaneous injection of carrageenan 24 hours

before behavioural testing. Signs of inflammation (redness, hypersensitivity and swelling)

were present at the hind paws of all animals. The paw withdrawal latency to thermal stimuli

was significantly decreased from 11.82±0.60 s to 8.34±0.51 s (n=12, p<0.05, paired t-test).

The control sEPSC frequency 1.28±0.24 Hz (n=56) recorded in neurons 1 day after the

inflammation induction was higher but not statistically different when compared to the control

sEPSC frequency recorded in naive animals.

Application of 20:4-NAPE (20 µM, 4min) to slices prepared from the spinal cord of these

animals strongly inhibited the sEPSC frequency in 7 of the 9 recorded neurons (59.5±15.6%,

n=9; Fig. 5). This 20:4-NAPE-induced inhibitory effect on frequency of sEPSC under

inflammatory conditions was not significantly different from that observed in naive rats

(52.3±7.8%, n=13; Fig. 2B). Application of 20:4-NAPE also reduced the amplitude of

sEPSCs from 21.4±2.3 pA to 18.4±1.6 pA (n=9, p=0.05, paired t-test). However, decrease of

more than 15% was present only in 3 from 9 recorded neurons.

Page 24 of 76



For Peer Review

13

The inhibitory effect of 20:4-NAPE on sEPSCs was mediated through the CB1 with

TRPV1 receptors participation under inflammatory conditions

Blocking the CB1 and TRPV1 receptors was tested on the 20:4-NAPE-induced inhibitory

effect in slices after the induction of peripheral inflammation. The CB1 receptor antagonist

PF514273 (0.2 µM, 6min) pretreatment did not induce any change in the sEPSC frequency

(108.0±9.8%, n=16; Fig. 6A,C). Subsequent co-application of PF514237 (0.2 µM) and 20:4-

NAPE (20 µM) did not change significantly the sEPSCs frequency when compared to the

control value (148.8±16.8%, n=16; Fig. 6A,C). In 9 of these 16 recorded neurons the

frequency of sEPSC increased, in 4 neurons it did not change and it decreased in 3 neurons.

The amplitude of sEPSCs was not significantly affected by PF514273 application alone

(control: 27.4±2.4 pA, PF514273: 24.6±2.5 pA) or PF514273 with 20:4-NAPE co-application

(24.1±2.2 pA, n=16).

The application of the TRPV1 antagonist SB366791 (10 µM, 6min) significantly decreased

the frequency of sEPSCs (71.5±10.9%, n=16; Fig. 6B,D). Subsequent co-application of

SB366791 (10 µM) and 20:4-NAPE (20 µM) induced a further decrease of the sEPSC

frequency compared to the control value (55.2±12.9%; Fig. 6B,D). When responses of the

individual cells were assessed, 11 of the 16 neurons exhibited decrease in sEPSC frequency

and it did not change in the rest of the cells. The amplitude of sEPSCs was not significantly

changed during the entire experiment (control: 24.2±1.7 pA, SB366791: 24.2±1.6 pA,

SB366791 + 20:4-NAPE: 21.8±1.2 pA, n=16).

The effect of combined inhibition of both receptors was evaluated in further experiments.

Pretreatment with PF514273 (0.2 µM) and SB366791 (10 µM) did not significantly change

the sEPSC frequency when all the neurons were pooled together (99.6±12.7%, n=15; Fig.

6E). Although in 9 of these 15 recorded neurons antagonists co-application decreased the

sEPSC frequency (64.9±4.8%, p<0.05, RM ANOVA on ranks followed by Student-Newman-

Keuls test), in 5 neurons it increased the sEPSC frequency (160.6±11.2%, p>0.05, RM

ANOVA on ranks). Subsequent co-application of PF514273 (0.2 µM), SB366791 (10 µM)

and 20:4-NAPE (20 µM) decreased the sEPSC frequency (76.9±11.6%, n=15; Fig. 6E)

compared to the control values. Antagonist co-treatment prevented the inhibitory effect of

20:4-NAPE application in 7 of the 15 cells. The amplitude of sEPSCs was not significantly

Page 25 of 76



For Peer Review

14

changed during different recording conditions (control: 24.9±3.7 pA, SB366791 + PF514273:

23.9±3.2 pA, SB366791 + PF514273 + 20:4-NAPE: 20.0±2.8 pA, n=15).

In addition, in order to compare the overall effect of the 20:4-NAPE application under the

different experimental conditions, the data were analysed also as a percentage of the previous

condition (=100%) and the differences statistically evaluated (Fig. 6F). Under this assessment

the sEPSC frequency after the 20:4-NAPE application alone was 59.5±15.6% of the control

(n=9). The sEPSC frequency after PF514237 + 20:4-NAPE application as a percentage of the

PF514273 pretreatment was significantly increased (152.7±21.4%, n=16), the sEPSC

frequency during the SB366791 + 20:4-NAPE application as a percentage of SB366791

pretreatment was (77.8±10.6%, n=16) and PF514237 + SB366791 + 20:4-NAPE as a

percentage of the PF514237 + SB366791 pretreatment was 81.2±8.2% (n=15). The increase

after the PF514237 + 20:4-NAPE application was significantly different from all the other

conditions.

These results suggest that the inhibitory effect induced by 20:4-NAPE application on the

sEPSC frequency is preferentially mediated by activation of CB1 receptors under the

inflammatory conditions (Fig. 6F). Moreover, when CB1 receptors were blocked the 20:4-

NAPE application led to increased sEPSC frequency, that was prevented by TRPV1 receptor

inhibition.

The reduction of the eEPSC amplitude induced by application of 20:4-NAPE was

prevented by blocking either the CB1 or TRPV1 receptors under the inflammatory

conditions

The 20:4-NAPE (20 µM) application during recording of eEPSCs in dorsal horn neurons after

dorsal root stimulation in spinal cord slices prepared 24 h after induction of peripheral

inflammation, significantly decreased the eEPSCs amplitude (78.5±6.6%, n=14; Fig. 7). This

decrease (>15%) was present in 9 of the 14 recorded neurons.

Treatment of the slices with CB1 receptor antagonist PF514273 (0.2 µM, 6min) did not

change the eEPSCs amplitude (109.0±13.6%, n=16, Fig. 8A,C). Subsequent co-application of

PF514273 (0.2 µM, 4min) and 20:4-NAPE (20 µM) increased the amplitude of eEPSCs

without reaching statistical significance (125.1±26.6%, Fig. 8A,C) when compared to the

Page 26 of 76



For Peer Review

15

control values. A CB1 receptor antagonist thus prevented the inhibitory effect induced by

application of 20:4-NAPE in 11 of 16 neurons.

Application of the TRPV1 antagonist SB366791 (10 µM, 6min) in another group of neurons

did not change the eEPSC amplitude either (88.0±10.8%, n=11, Fig. 8B,D). Subsequent co-

application of SB366791 (10 µM, 4min) and 20:4-NAPE (20 µM) did not change the

amplitude of eEPSC (92.0±17.8%, Fig. 8B,D) compared to the control values. Similarly to the

inhibition of CB1 receptors, the TRPV1 receptor antagonist prevented the 20:4-NAPE-

induced inhibitory effect in majority (8 of 11) of the recorded superficial dorsal horn neurons.

In addition, to compare the 20:4-NAPE effect between the different experimental conditions,

the same data were expressed as a percentage of the previous application: 20:4-NAPE as

percentage of eEPSC basal amplitude (78.5±6.6%, n=14); PF514237 + 20:4-NAPE as a

percentage of PF514273 pretreatment (117.7±17.7%, n=16); SB366791 + 20:4-NAPE as a

percentage of SB366791 pretreatment (101.3±16.0%, n=11, Fig. 8E). These results indicate

that under the inflammatory conditions both CB1 and TRPV1 receptors mediated the

inhibitory effect induced by the 20:4-NAPE application on evoked EPSC amplitude.

Discussion

Here we report that 20:4-NAPE application induced an inhibitory effect on excitatory

nociceptive synaptic transmission demonstrated by decrease of sEPSC frequency and

reduction of dorsal root stimulation-evoked EPSC amplitude in the superficial spinal dorsal

horn. The inhibitory effect occurred both in naive conditions and following the development

of hindpaw inflammation. The differential effects of CB1 and TRPV1 antagonists indicated

that the underlying mechanisms of 20:4-NAPE-induced inhibition may differ in those two

conditions.

20:4-NAPE and anandamide synthesis

20:4-NAPE is a substrate for anandamide synthesis in enzyme preparations and cultured

primary sensory neurons (Varga et al., 2014; Wang et al., 2006). Here we found that spinal

cord slices also produce AEA after 20:4-NAPE application. Although direct effects of 20:4-

NAPE, or indirect effects through a metabolite other than anandamide on some other

Page 27 of 76



For Peer Review

16

receptors cannot be categorically excluded, we propose that at least the great majority of the

20:4-NAPE application-induced effects described here, was mediated through the synthesis of

anandamide acting on CB1 and TRPV1 receptors.

Anandamide activation of CB1 and TRPV1 receptors

The CB1 receptor and TRPV1 constitute the main targets for anandamide (Devane et al.,

1992; Zygmunt et al., 1999) and our data show that those two receptors mediate at least the

majority of the 20:4-NAPE application-induced effects. However, as anandamide is a highly

promiscuous molecule, the involvement of other molecules including peroxisome proliferator-

activated receptor alpha and gamma, sodium and T-type Ca2+

channels (Kim et al., 2005;

Okura et al., 2014; Chemin et al., 2001; O'Sullivan, 2007) cannot be ruled out. Nevertheless,

due to the robust effects of the CB1 receptor and TRPV1 antagonists, here we addressed the

contribution of these two receptors only.

The 20:4-NAPE application-induced inhibitory effect on sEPSC frequency and eEPSC

amplitude was mediated preferentially by activation of CB1 receptor under naive conditions.

Although post-synaptic CB1 receptor expression in the spinal cord has been reported

(Farquhar-Smith et al., 2000), most studies suggest exclusive pre-synaptic location either on

DRG neuron terminals or terminals of GABAergic inhibitory interneurons (Hegyi et al.,

2012; Nyilas et al., 2009; Pernia-Andrade et al., 2009; Veress et al., 2013). CB1 receptor

activation at both locations leads to reduced transmitter release (Morisset et al., 2001; Nyilas

et al., 2009; Pernia-Andrade et al., 2009). In our preparations the inhibitory synaptic

transmission was pharmacologically blocked. Therefore, it seems plausible to suggest that the

CB1 receptor-mediated inhibitory effect by 20:4-NAPE application occurred through

anandamide-mediated CB1 receptor activation and subsequent reduction of transmitter release

from spinal terminals of DRG neurons.

Under the naive condition, there was a tendency of CB1 receptors antagonist per se to mildly

increase the frequency of sEPSC, although this increase did not reach statistical significance.

TRPV1 antagonist did not have any effect on the superficial dorsal horn neurons sEPSC

frequency, similar to our previous experiments (Spicarova et al., 2014a). Nevertheless

moderate TRPV1 mediated sEPSC tonic activity was reported in lamina II neurons in mice

(Park et al., 2011).

Page 28 of 76



For Peer Review

17

The robust decrease of AMPA receptor-mediated sEPSC frequency (Spicarova and Palecek,

2010) induced by 20:4-NAPE application was also accompanied by moderate reduction of

sEPSC amplitude in some neurons. In our preparation the recorded neurons have contacts

with numerous synapses, which spontaneously release glutamate and induce sEPSCs. The

robust decrease of sEPSC frequency could elicit strong attenuation of glutamate release from

specific, 20:4-NAPE application-responding afferents leading to average sEPSC amplitude

decrease without affecting the postsynaptic mechanisms.

The effect of peripheral inflammation

In slices taken after the induction of peripheral inflammation, 20:4-NAPE application induced

a significant inhibition of sEPSC frequency similar to the naive preparations. However,

SB366791 reduced sEPSC frequency, suggesting presynaptic TRPV1 receptors tonic

activation. The effect of SB366971 per se is consistent with inflammation-induced tonic

activity (Lappin et al., 2006) and increased sensitivity to endogenous agonists (Spicarova and

Palecek, 2009) of presynaptic TRPV1 in the spinal cord dorsal horn. TRPV1 is expressed in

the overwhelming majority of spinal C-fibre terminals in the superficial dorsal horn (Caterina

et al., 1997; Guo et al., 1999). Consistently with this high TRPV1 expression, regulation

(activation, desensitisation and inhibition) of TRPV1 has large impact on glutamate release

from these afferents (Spicarova et al., 2014b). It was suggested that modulation of TRPV1 in

the dorsal horn could underlie several pathological pain states (Kanai et al., 2005; Spicarova

et al., 2014a; Spicarova et al., 2011).

Tonic activation of presynaptic CB1 receptors was not detected under the inflammatory

conditions. However, the CB1 receptor antagonist prevented the 20:4-NAPE application-

produced inhibitory effect on sEPSC frequency. Moreover, 20:4-NAPE application

significantly increased the frequency of sEPSCs, when CB1 receptors were blocked and this

potentiating effect was prevented by TRPV1 receptor inhibition (Fig. 6F). This indicates that

under inflammatory conditions 20:4-NAPE application-induced inhibition of the sEPSC

frequency was mediated by CB1 receptors while the potentiating effect mediated by TRPV1

receptors was unmasked only when CB1 receptors were blocked.

The CB1 receptor-mediated block of 20:4-NAPE application-induced inhibitory effect on

eEPSC amplitude was maintained after the development of inflammation. However, this 20:4-

NAPE application-induced inhibitory effect was prevented by blocking either CB1 or TRPV1

Page 29 of 76



For Peer Review

18

receptors, indicating involvement of both receptors. We did not observe a significant

reduction of eEPSC amplitude after TRPV1 antagonist application as with the sEPSC. While

it is possible that activation of TRPV1 receptors under these conditions did not play such an

important role, it needs also to be taken into account that the electrical stimulation of dorsal

root could activate also myelinated primary afferents that do not express TRPV1 receptors

(Caterina et al., 1997; Guo et al., 1999). The effect of the TRPV1 antagonist application thus

could be “diluted”.

In contrast to potentiation of the spontaneous transmitter release by TRPV1 agonist, the

release induced by action potentials evoked by dorsal root electrical stimulation may be

blocked by TRPV1 receptors activation (Yang et al., 1999, Baccei et al., 2003). Thus it is

plausible, that activation of TRPV1 on presynaptic terminals of DRG neurons by 20:4-NAPE

application reduced the glutamate release from primary afferents and thus contributed to the

decrease of evoked EPSC amplitude in the recorded postsynaptic neuron. In addition, rapid

internalization of voltage activated Ca2+

channels by TRPV1 activation (Wu et al., 2005)

could underlie the reduction of synchronous transmitter release. Although, the vast majority

of spinal TRPV1 is localized in terminals of primary sensory neurons, postsynaptic TRPV1

expression was also described in some GABAergic neurons, in which TRPV1 activation

induces long-term depression through the reduction of AMPA channels in the plasmatic

membrane (Caterina et al., 1997; Guo et al., 1999; Kim et al., 2012). We cannot exclude the

possibility that our neurons recorded in laminae I and II(outer) could include GABAergic cells

in which the postsynaptic TRPV1-mediated modulation under the inflammatory conditions

could occur, though it would change only the EPSC amplitude.

The role of 20:4-NAPE and anandamide in nociceptive modulation

In summary, our data together indicate that 20:4-NAPE application induces mainly CB1

receptor mediated inhibitory effects on excitatory transmission in naive animals while TRPV1

mediated mechanisms are also involved after peripheral inflammation. We propose, that if the

20:4-NAPE application-induced effects are indeed mediated through anandamide synthesis,

balanced signalling by anandamide and its targets are involved in preventing the spread of

nociceptive signals into supraspinal structures and this balance may be compromised during

inflammation.

Page 30 of 76



For Peer Review

19

Anandamide, due to its lipophilic nature, is expected to be produced in close proximity to its

target. TRPV1-expressing primary sensory neurons indeed express multiple anandamide-

synthesising pathways (Sousa-Valente et al., 2014b; Sousa-Valente et al., 2017; Varga et al.,

2014). Further, transcripts of several anandamide-synthesising enzymes are expressed in the

spinal dorsal horn (Malek et al., 2014). Enhanced activity after inflammation and during the

electrical stimulation of primary afferent fibres resulted in increased concentration of Ca2+

in

presynaptic terminals and could induce or increase the enzymatic activity of NAPE-PLD.

Furthermore Ca2+

influx through postsynaptic AMPA receptors could be also involved

through promoting anandamide synthesis from 20:4-NAPE, as NAPE-PLD is expressed in

post-synaptic dendrites in the spinal dorsal horn (Hegyi et al., 2012). Ca2+

-insensitive

pathways and NAPE-PLD activity as a part of a retrograde inhibitory mechanism (Katona and

Freund, 2008) could be also involved in this anandamide synthesis. NAPE-PLD and other

anandamide-synthesising enzymes may be a particularly important for regulating nociceptive

spinal processing under inflammatory conditions.

The 20:4-NAPE application in our experiments also provided a distinctive opportunity to

study the role of the spinal endocannabinoid system, by application of substrate for

anandamide synthesis instead of anandamide directly. By this approach, physiological

mechanisms of anandamide synthesis played an important role, including the level of their

activity and local distribution, creating anandamide microdomains concentrations. By

flooding the preparation by anandamide directly, most likely other receptors and biological

pathways would have been activated. This method of local “on demand” anandamide

production from its precursor may prove to be of advantage also in the clinical settings for

pain treatment. Especially, as clinical trials focused to increase anandamide levels by reduced

hydrolysis with fatty acid amide hydrolase inhibitors did not show clinical efficacy (Mallet et

al., 2016).

Page 31 of 76



For Peer Review

20

Acknowledgements

This work was supported by GACR 15-11138S, MSMT LH15279, CZ.1.05/1.1.00/02.0109,

RVO67985823, GAUK138215. Authors would like to thank Prof. Ivan Miksik for the AEA

mass spectrometry analysis.

Author contributions

JP conceived and designed the study. VN, PM and PA conducted experiments, VN, PM and

DS analysed the data. VN, IN, DS and JP participated in writing the manuscript. All authors

read and approved the final version of the manuscript.

Conflict of interest

The authors declare no conflicts of interest.

Declaration of transparency and scientific rigour

This Declaration acknowledges that this paper adheres to the principles for transparent

reporting and scientific rigour of preclinical research recommended by funding agencies,

publishers and other organisations engaged with supporting research.

Page 32 of 76



For Peer Review

21

References

Ahluwalia J, Urban L, Bevan S, Nagy I (2003). Anandamide regulates neuropeptide release from

capsaicin-sensitive primary sensory neurons by activating both the cannabinoid 1 receptor and the

vanilloid receptor 1 in vitro. The European journal of neuroscience 17(12): 2611-2618.

Ahluwalia J, Urban L, Capogna M, Bevan S, Nagy I (2000). Cannabinoid 1 receptors are expressed in

nociceptive primary sensory neurons. Neuroscience 100(4): 685-688.

Alkaitis MS, Solorzano C, Landry RP, Piomelli D, DeLeo JA, Romero-Sandoval EA (2010). Evidence for a

role of endocannabinoids, astrocytes and p38 phosphorylation in the resolution of postoperative

pain. PloS one 5(5): e10891.

Amaya F, Oh-hashi K, Naruse Y, Iijima N, Ueda M, Shimosato G, et al. (2003). Local inflammation

increases vanilloid receptor 1 expression within distinct subgroups of DRG neurons. Brain research

963(1-2): 190-196.

Amaya F, Shimosato G, Kawasaki Y, Hashimoto S, Tanaka Y, Ji RR, et al. (2006). Induction of CB1

cannabinoid receptor by inflammation in primary afferent neurons facilitates antihyperalgesic effect

of peripheral CB1 agonist. Pain 124(1-2): 175-183.

Baccei ML, Bardoni R, Fitzgerald M (2003). Development of nociceptive synaptic inputs to the

neonatal rat dorsal horn: glutamate release by capsaicin and menthol. The Journal of physiology

549(Pt 1): 231-242.

Buczynski MW, Svensson CI, Dumlao DS, Fitzsimmons BL, Shim JH, Scherbart TJ, et al. (2010).

Inflammatory hyperalgesia induces essential bioactive lipid production in the spinal cord. Journal of

neurochemistry 114(4): 981-993.

Carrier EJ, Kearn CS, Barkmeier AJ, Breese NM, Yang W, Nithipatikom K, et al. (2004). Cultured rat

microglial cells synthesize the endocannabinoid 2-arachidonylglycerol, which increases proliferation

via a CB2 receptor-dependent mechanism. Molecular pharmacology 65(4): 999-1007.

Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD, Julius D (1997). The capsaicin

receptor: a heat-activated ion channel in the pain pathway. Nature 389(6653): 816-824.

Chemin J, Monteil A, Perez-Reyes E, Nargeot J, Lory P (2001). Direct inhibition of T-type calcium

channels by the endogenous cannabinoid anandamide. The EMBO journal 20(24): 7033-7040.

Costa B, Bettoni I, Petrosino S, Comelli F, Giagnoni G, Di Marzo V (2010). The dual fatty acid amide

hydrolase/TRPV1 blocker, N-arachidonoyl-serotonin, relieves carrageenan-induced inflammation and

hyperalgesia in mice. Pharmacological research : the official journal of the Italian Pharmacological

Society 61(6): 537-546.

Curtis MJ, Bond RA, Spina D, Ahluwalia A, Alexander SP, Giembycz MA, et al. (2015). Experimental

design and analysis and their reporting: new guidance for publication in BJP. British journal of

pharmacology 172(14): 3461-3471.

Devane WA, Hanus L, Breuer A, Pertwee RG, Stevenson LA, Griffin G, et al. (1992). Isolation and

structure of a brain constituent that binds to the cannabinoid receptor. Science 258(5090): 1946-

1949.

Dow RL, Carpino PA, Hadcock JR, Black SC, Iredale PA, DaSilva-Jardine P, et al. (2009). Discovery of 2-

(2-chlorophenyl)-3-(4-chlorophenyl)-7-(2,2-difluoropropyl)-6,7-dihydro-2H-pyraz olo[3,4-

Page 33 of 76



For Peer Review

22

f][1,4]oxazepin-8(5H)-one (PF-514273), a novel, bicyclic lactam-based cannabinoid-1 receptor

antagonist for the treatment of obesity. Journal of medicinal chemistry 52(9): 2652-2655.

Farquhar-Smith WP, Egertova M, Bradbury EJ, McMahon SB, Rice AS, Elphick MR (2000). Cannabinoid

CB(1) receptor expression in rat spinal cord. Molecular and cellular neurosciences 15(6): 510-521.

Gunthorpe MJ, Rami HK, Jerman JC, Smart D, Gill CH, Soffin EM, et al. (2004). Identification and

characterisation of SB-366791, a potent and selective vanilloid receptor (VR1/TRPV1) antagonist.

Neuropharmacology 46(1): 133-149.

Guo A, Vulchanova L, Wang J, Li X, Elde R (1999). Immunocytochemical localization of the vanilloid

receptor 1 (VR1): relationship to neuropeptides, the P2X3 purinoceptor and IB4 binding sites. The

European journal of neuroscience 11(3): 946-958.

Hegyi Z, Hollo K, Kis G, Mackie K, Antal M (2012). Differential distribution of diacylglycerol lipase-

alpha and N-acylphosphatidylethanolamine-specific phospholipase d immunoreactivity in the

superficial spinal dorsal horn of rats. Glia 60(9): 1316-1329.

Kanai Y, Nakazato E, Fujiuchi A, Hara T, Imai A (2005). Involvement of an increased spinal TRPV1

sensitization through its up-regulation in mechanical allodynia of CCI rats. Neuropharmacology 49(7):

977-984.

Katona I, Freund TF (2008). Endocannabinoid signaling as a synaptic circuit breaker in neurological

disease. Nature medicine 14(9): 923-930.

Kilkenny C, Browne W, Cuthill IC, Emerson M, Altman DG (2010). Animal research: reporting in vivo

experiments: the ARRIVE guidelines. British journal of pharmacology 160(7): 1577-1579.

Kim YH, Back SK, Davies AJ, Jeong H, Jo HJ, Chung G, et al. (2012). TRPV1 in GABAergic interneurons

mediates neuropathic mechanical allodynia and disinhibition of the nociceptive circuitry in the spinal

cord. Neuron 74(4): 640-647.

Kim HI, Kim TH, Shin YK, Lee CS, Park M, Song JH (2005). Anandamide suppression of Na+ currents in

rat dorsal root ganglion neurons. Brain research 1062(1-2): 39-47.

Kwon SG, Roh DH, Yoon SY, Moon JY, Choi SR, Choi HS, et al. (2014). Blockade of peripheral P2Y1

receptors prevents the induction of thermal hyperalgesia via modulation of TRPV1 expression in

carrageenan-induced inflammatory pain rats: involvement of p38 MAPK phosphorylation in DRGs.

Neuropharmacology 79: 368-379.

Lappin SC, Randall AD, Gunthorpe MJ, Morisset V (2006). TRPV1 antagonist, SB-366791, inhibits

glutamatergic synaptic transmission in rat spinal dorsal horn following peripheral inflammation.

European journal of pharmacology 540(1-3): 73-81.

Lever IJ, Robinson M, Cibelli M, Paule C, Santha P, Yee L, et al. (2009). Localization of the

endocannabinoid-degrading enzyme fatty acid amide hydrolase in rat dorsal root ganglion cells and

its regulation after peripheral nerve injury. The Journal of neuroscience : the official journal of the

Society for Neuroscience 29(12): 3766-3780.

Luo H, Cheng J, Han JS, Wan Y (2004). Change of vanilloid receptor 1 expression in dorsal root

ganglion and spinal dorsal horn during inflammatory nociception induced by complete Freund's

adjuvant in rats. Neuroreport 15(4): 655-658.

Malek N, Kucharczyk M, Starowicz K (2014). Alterations in the anandamide metabolism in the

development of neuropathic pain. BioMed research international 2014: 686908.

Page 34 of 76



For Peer Review

23

Mallet C, Dubray C, Duale C (2016). FAAH inhibitors in the limelight, but regrettably. International

journal of clinical pharmacology and therapeutics 54(7): 498-501.

McGrath JC, Lilley E (2015). Implementing guidelines on reporting research using animals (ARRIVE

etc.): new requirements for publication in BJP. British journal of pharmacology 172(13): 3189-3193.

Morisset V, Ahluwalia J, Nagy I, Urban L (2001). Possible mechanisms of cannabinoid-induced

antinociception in the spinal cord. European journal of pharmacology 429(1-3): 93-100.

Morisset V, Urban L (2001). Cannabinoid-induced presynaptic inhibition of glutamatergic EPSCs in

substantia gelatinosa neurons of the rat spinal cord. Journal of neurophysiology 86(1): 40-48.

Nagy I, Friston D, Valente JS, Torres Perez JV, Andreou AP (2014). Pharmacology of the capsaicin

receptor, transient receptor potential vanilloid type-1 ion channel. Progress in drug research.

Fortschritte der Arzneimittelforschung. Progres des recherches pharmaceutiques 68: 39-76.

Nyilas R, Gregg LC, Mackie K, Watanabe M, Zimmer A, Hohmann AG, et al. (2009). Molecular

architecture of endocannabinoid signaling at nociceptive synapses mediating analgesia. The


Okura D, Horishita T, Ueno S, Yanagihara N, Sudo Y, Uezono Y, et al. (2014). The endocannabinoid

anandamide inhibits voltage-gated sodium channels Nav1.2, Nav1.6, Nav1.7, and Nav1.8 in Xenopus

oocytes. Anesthesia and analgesia 118(3): 554-562.

O'Sullivan SE (2007). Cannabinoids go nuclear: evidence for activation of peroxisome proliferator-

activated receptors. British journal of pharmacology 152(5): 576-582.

Park CK, Lu N, Xu ZZ, Liu T, Serhan CN, Ji RR (2011). Resolving TRPV1- and TNF-alpha-mediated spinal

cord synaptic plasticity and inflammatory pain with neuroprotectin D1. The Journal of neuroscience :

the official journal of the Society for Neuroscience 31(42): 15072-15085.

Pernia-Andrade AJ, Kato A, Witschi R, Nyilas R, Katona I, Freund TF, et al. (2009). Spinal

endocannabinoids and CB1 receptors mediate C-fiber-induced heterosynaptic pain sensitization.

Science 325(5941): 760-764.

Ren K, Dubner R (1999). Inflammatory Models of Pain and Hyperalgesia. ILAR journal 40(3): 111-118.

Richardson JD, Kilo S, Hargreaves KM (1998). Cannabinoids reduce hyperalgesia and inflammation via

interaction with peripheral CB1 receptors. Pain 75(1): 111-119.

Snider NT, Walker VJ, Hollenberg PF (2010). Oxidation of the endogenous cannabinoid arachidonoyl

ethanolamide by the cytochrome P450 monooxygenases: physiological and pharmacological

implications. Pharmacological reviews 62(1): 136-154.

Sousa-Valente J, Andreou AP, Urban L, Nagy I (2014a). Transient receptor potential ion channels in

primary sensory neurons as targets for novel analgesics. British journal of pharmacology 171(10):

2508-2527.

Sousa-Valente J, Varga A, Ananthan K, Khajuria A, Nagy I (2014b). Anandamide in primary sensory

neurons: too much of a good thing? The European journal of neuroscience 39(3): 409-418.

Sousa-Valente J, Varga A, Torres Perez JV, Jenes A, Wahba J, Mackie K, et al. (2017). Inflammation of

peripheral tissues and injury to peripheral nerves induce diferring effects in the expression of the

calcium-sensitive anandamide-synthesising enzyme and related molecules in rat primary sensory

neurons. The Journal of comparative neurology.

Page 35 of 76



For Peer Review

24

Spicarova D, Adamek P, Kalynovska N, Mrozkova P, Palecek J (2014a). TRPV1 receptor inhibition

decreases CCL2-induced hyperalgesia. Neuropharmacology 81: 75-84.

Spicarova D, Nerandzic V, Palecek J (2011). Modulation of spinal cord synaptic activity by tumor

necrosis factor alpha in a model of peripheral neuropathy. Journal of neuroinflammation 8: 177.

Spicarova D, Nerandzic V, Palecek J (2014b). Update on the role of spinal cord TRPV1 receptors in

pain modulation. Physiological research / Academia Scientiarum Bohemoslovaca 63 Suppl 1: S225-

236.

Spicarova D, Palecek J (2010). Modulation of AMPA excitatory postsynaptic currents in the spinal

cord dorsal horn neurons by insulin. Neuroscience 166(1): 305-311.

Spicarova D, Palecek J (2009). The role of the TRPV1 endogenous agonist N-Oleoyldopamine in

modulation of nociceptive signaling at the spinal cord level. Journal of neurophysiology 102(1): 234-

243.

Tognetto M, Amadesi S, Harrison S, Creminon C, Trevisani M, Carreras M, et al. (2001). Anandamide

excites central terminals of dorsal root ganglion neurons via vanilloid receptor-1 activation. The

Journal of neuroscience : the official journal of the Society for Neuroscience 21(4): 1104-1109.

Ueda N, Tsuboi K, Uyama T (2013). Metabolism of endocannabinoids and related N-

acylethanolamines: canonical and alternative pathways. The FEBS journal 280(9): 1874-1894.

van der Stelt M, Trevisani M, Vellani V, De Petrocellis L, Schiano Moriello A, Campi B, et al. (2005).

Anandamide acts as an intracellular messenger amplifying Ca2+ influx via TRPV1 channels. The EMBO

journal 24(17): 3026-3037.

Varga A, Jenes A, Marczylo TH, Sousa-Valente J, Chen J, Austin J, et al. (2014). Anandamide produced

by Ca(2+)-insensitive enzymes induces excitation in primary sensory neurons. Pflugers Archiv :

European journal of physiology 466(7): 1421-1435.

Vellani V, Petrosino S, De Petrocellis L, Valenti M, Prandini M, Magherini PC, et al. (2008). Functional

lipidomics. Calcium-independent activation of endocannabinoid/endovanilloid lipid signalling in

sensory neurons by protein kinases C and A and thrombin. Neuropharmacology 55(8): 1274-1279.

Veress G, Meszar Z, Muszil D, Avelino A, Matesz K, Mackie K, et al. (2013). Characterisation of

cannabinoid 1 receptor expression in the perikarya, and peripheral and spinal processes of primary

sensory neurons. Brain structure & function 218(3): 733-750.

Wang J, Okamoto Y, Morishita J, Tsuboi K, Miyatake A, Ueda N (2006). Functional analysis of the

purified anandamide-generating phospholipase D as a member of the metallo-beta-lactamase family.

The Journal of biological chemistry 281(18): 12325-12335.

Wang J, Ueda N (2009). Biology of endocannabinoid synthesis system. Prostaglandins & other lipid

mediators 89(3-4): 112-119.

Wu ZZ, Chen SR, Pan HL (2005). Transient receptor potential vanilloid type 1 activation down-

regulates voltage-gated calcium channels through calcium-dependent calcineurin in sensory neurons.


Yang K, Kumamoto E, Furue H, Li YQ, Yoshimura M (1999). Action of capsaicin on dorsal root-evoked

synaptic transmission to substantia gelatinosa neurons in adult rat spinal cord slices. Brain research

830(2): 268-273.

Page 36 of 76



For Peer Review

25

Zygmunt PM, Petersson J, Andersson DA, Chuang H, Sorgard M, Di Marzo V, et al. (1999). Vanilloid

receptors on sensory nerves mediate the vasodilator action of anandamide. Nature 400(6743): 452-

457.

Page 37 of 76



For Peer Review

26

Figure 1. Anandamide concentration after 20:4-NAPE application in spinal cord slices.

Three different concentrations of 20:4-NAPE (20 µM, 100 µM, 200 µM) were applied on

spinal cord slices. Increasing content of AEA was detected in the extracellular solution after

20:4-NAPE application in a concentration dependent manner (n=5, *p<0.05 compared to

control, #p<0.05 compared to 20 µM and 100 µM 20:4-NAPE application, RM ANOVA on

ranks followed by Student-Newman-Keuls test).

Figure 2. Inhibitory effect of 20:4-NAPE application on excitatory postsynaptic currents

in spinal cord slices from naive animals. (A) An example of native recording of

spontaneous EPSCs from one superficial dorsal horn neuron before (CTRL) and during 20:4-

NAPE (20 µM) application. (B) Application of 20:4-NAPE (20 µM) robustly decreased the

average frequency of sEPSCs (n=13, *p<0.05, Wilcoxon signed-rank test) (C) This is also

evident using cumulative histogram analysis (p<0.05, Kolmogorov-Smirnov test). (D)

Decrease of sEPSC amplitude was not significant using cumulative amplitude analysis. (E)

Recording of dorsal root stimulation-evoked EPSC from one neuron before and during 20:4-

NAPE (20 µM) application. (F) Acute application of 20:4-NAPE (20 µM) significantly

decreased the mean amplitude of eEPSCs (n=15,*p<0.05, Wilcoxon signed-rank test).

Figure 3. The effect of CB1 and TRPV1 receptor antagonists on the 20:4-NAPE-induced

inhibition of sEPSC frequency in naive slices. (A, C) The application of PF514273 (0.2

µM) alone did not change the frequency of sEPSCs significantly (n=11). Following co-

application of PF514273 (0.2 µM) with 20:4-NAPE (20 µM) prevented the 20:4-NAPE

induced inhibition and the sEPSCs frequency did not differ from the control. (B, D) The

application of SB366791 (10 µM, n=10) did not change the sEPSCs frequency. However,

following co-application of SB366791 (10 µM) and 20:4-NAPE (20 µM) a significant

decrease in the sEPSCs frequency was present (*p<0.05 versus control, #p<0.05 versus

SB366791 pretreatment, RM ANOVA on ranks followed by Student-Newman-Keuls test).

(E) The application of both antagonists PF514273 (0.2 µM) with SB366791 (10 µM) did not

change the frequency of sEPSCs significantly (n=8). Subsequent co-application of PF514273

(0.2 µM), SB366791 (10 µM) with 20:4-NAPE (20 µM) prevented the 20:4-NAPE induced

inhibition. (F) The same data are expressed as a percentage of previous recording conditions:

20:4-NAPE (n=13, *p<0.05, Wilcoxon signed-rank test) versus basal frequency of sEPSCs;

PF514273 + 20:4-NAPE (n=11) versus the PF514273 pretreatment; SB366791 + 20:4-NAPE

(n=10, *p<0.05, Wilcoxon signed-rank test) versus the SB366791 pretreatment; PF514237 +

Page 38 of 76



For Peer Review

27

SB366791 + 20:4-NAPE compared to both antagonists pretreatment (n=8). Statistically

significant difference between 20:4-NAPE alone and PF514273 + 20:4-NAPE: #p<0.05, One-

way ANOVA followed by Student-Newman-Keuls test.

Figure 4. Effect of CB1 and TRPV1 antagonists on 20:4-NAPE-induced inhibition of

eEPSC amplitude in naive slices. (A, C) The pretreatment with PF514273 (0.2 µM, n=13)

did not changed the amplitude of the recorded eEPSC in spinal cord slices prepared from

naive animals. Following co-application of PF514273 (0.2 µM) and 20:4-NAPE (20 µM) also

did not significantly changed the amplitude of eEPSC. (B, D) The pretreatment with

SB366791 (10 µM) elicited very similar size of the control eEPSC amplitude. Following co-

application of SB366791 (10 µM) and 20:4-NAPE (20 µM) induced decrease of the eEPSC

amplitude (n=10, *p<0.05 versus control, #p<0.05 versus SB366791 pretreatment, RM

ANOVA on ranks followed by Student-Newman-Keuls test). (E) Data are shown as a

percentage of previous condition to eliminate the effect of antagonist activity, respectively:

20:4-NAPE as percentage of eEPSC basal amplitude (n=15, *p<0.05, Wilcoxon signed-rank

test); PF514273 + 20:4-NAPE as a percentage of PF514273 pretreatment (n=13); SB366791 +

20:4-NAPE as a percentage of SB366791 pretreatment (n=10, *p<0.05, Wilcoxon signed-rank

test).

Figure 5. Application of the 20:4-NAPE decreased the frequency of sEPSCs under

inflammatory conditions. (A) Native recording from one superficial dorsal horn neuron

before and during 20:4-NAPE (20 µM) bath application on spinal cord slice dissected 24 h

after the induction of peripheral inflammation. (B) Application of 20:4-NAPE (20 µM)

significantly decreased the frequency of sEPSCs (n=9, *p<0.05, Wilcoxon signed-rank test).

Figure 6. The effect of CB1 and TRPV1 antagonists on 20:4-NAPE-induced inhibition of

sEPSC frequency under inflammatory conditions: (A, C) The application of PF514273

(0.2 µM, n=16) did not change the frequency of sEPSCs. Subsequent co-application of

PF514273 (0.2 µM) and 20:4-NAPE (20 µM) increased the frequency of sEPSCs without

statistical significance compared to control. (B, D) The frequency of sEPSCs significantly

decreased during application of SB366791 (10 µM, n=16, *p<0.05, RM ANOVA on ranks

followed by Student-Newman-Keuls test). Following co-application of SB366791 (10 µM)

and 20:4-NAPE (20 µM) leaded to even stronger decrease of sEPSC frequency (*p<0.05, RM

ANOVA on ranks followed by Student-Newman-Keuls test). (E) The combined application of

Page 39 of 76



For Peer Review

28

PF514273 (0.2 µM) and SB366791 (10 µM) did not change the frequency of sEPSCs,

subsequent co-application of both antagonists with 20:4-NAPE (20 µM) significantly

decreased the frequency of sEPSCs only compared to control (n=15, *p<0.05, RM ANOVA

on ranks followed by Student-Newman-Keuls test). (F) The same data are shown as a

percentage of previous recording conditions: 20:4-NAPE (n=9) versus sEPSC basal

frequency, PF514237 + 20:4-NAPE versus PF514273 (n=16) pretreatment, SB366791 + 20:4-

NAPE versus SB366791 (n=16) pretreatment and PF514237 + SB366791 + 20:4-NAPE

versus both antagonists pretreatment (n=15). Statistically significant differences: *p<0.05

versus pretreatment, Wilcoxon signed-rank test; #p<0.05 versus PF514237 + 20:4-NAPE co-

application, One-way ANOVA followed by Student-Newman-Keuls test.

Figure 7. Application of 20:4-NAPE decreased the amplitude of evoked EPSCs in

superficial dorsal horn neurons under inflammatory conditions. (A) An example of native

recording from one nociceptive neuron before and during 20:4-NAPE (20 µM) bath

application on acute spinal cord slice prepared 24 h after intraplantar injection of carrageenan.

(B) Acute application of 20:4-NAPE (20 µM) significantly decreased the amplitude of

eEPSCs (n=14; *p<0.05, Wilcoxon signed-rank test).

Figure 8. Antagonists of CB1 and TRPV1 receptors blocked the 20:4-NAPE-induced

decrease of eEPSC amplitude under inflammatory conditions. (A, C) The application of

PF514273 (0.2 µM, n=16) did not changed the amplitude of eEPSCs. Following co-

application of PF514273 (0.2 µM) and 20:4-NAPE (20 µM) slightly increase the amplitude of

eEPSCs without statistical significance. (B, D) The pretreatment with SB366791 (10 µM,

n=11) did not change the amplitude of eEPSC. Following co-application of SB366791 (10

µM) and 20:4-NAPE (20 µM) also did not change the eEPSC amplitude. (E) The same data

are shown as a percentage of previous recording conditions: 20:4-NAPE versus eEPSC basal

amplitude (n=14, *p<0.05, Wilcoxon signed-rank test); PF514237 + 20:4-NAPE versus

PF514273 pretreatment (n=16); SB366791 + 20:4-NAPE versus SB366791 pretreatment

(n=11).

Page 40 of 76



For Peer Review

1

Peripheral inflammation alters N-arachidonoylphosphatidylethanolamine (20:4-NAPE)

induced modulation of nociceptive spinal cord synaptic transmission

20:4-NAPE-modulated spinal synaptic transmission

Vladimir Nerandzic1, Petra Mrozkova

1, Pavel Adamek

1.3, Diana Spicarova

1,

Istvan Nagy2 and Jiri Palecek

1

1Department of Functional Morphology, Institute of Physiology,

The Czech Academy of Sciences, Prague, Czech Republic

2Department of Anaesthetics, Pain Medicine and Intensive Care, Imperial College London,

Faculty of Medicine, Chelsea and Westminster Hospital, London, UK

3Faculty of Science, Charles University, Prague, Czech Republic

Corresponding author:

Jiri Palecek M.D., Ph.D.

Department of Functional Morphology

Institute of Physiology Czech Academy of Sciences

Videnska 1083

142 20 Praha 4

Czech Republic

e-mail: [email protected]

office: (-420) 2 4106 2664

fax: (-420) 2 4106 2488

Page 41 of 76



For Peer Review

2

Abstract

Background and Purpose

Endocannabinoids play an important role in modulating spinal nociceptive signalling, crucial

for the development of pain. The cannabinoid receptor 1 (CB1) and the transient receptor

potential cation channel subfamily V member 1 (TRPV1) are both activated by the

endocannabinoid anandamide that is a product of biosynthesis from the endogenous lipid

precursor N-arachidonoylphosphatidylethanolamine (20:4-NAPE). Here we are first to report

CB1 receptor- and TRPV1-mediated effects of 20:4-NAPE application on spinal synaptic

transmission in control and inflammatory conditions.

Experimental Approach

Spontaneous (sEPSCs) and dorsal root stimulation-evoked (eEPSCs) excitatory postsynaptic

currents from superficial dorsal horn neurons in rat spinal cord slices were assessed.

Peripheral inflammation was induced by carrageenan. Anandamide concentration was

assessed by mass spectrometry.

Key Results

Application of 20:4-NAPE increased anandamide concentration in vitro. 20:4-NAPE (20 µM)

decreased sEPSCs frequency and eEPSCs amplitude in control and inflammatory conditions.

The inhibitory effect of 20:4-NAPE was sensitive to CB1 antagonist PF514273 (0.2 µM) in

both conditions, but to the TRPV1 antagonist SB366791 (10 µM) only after inflammation.

After inflammation 20:4-NAPE increased sEPSCs frequency in the presence of PF514273 and

this increase was blocked by SB366791.

Conclusions and Implications

While 20:4-NAPE treatment produced an inhibitory effect on excitatory synaptic transmission

in both naive and inflammatory conditions, peripheral inflammation altered the underlying

mechanisms. Our data indicate that 20:4-NAPE application induced mainly CB1 receptor-

mediated inhibitory effects in naive animals while TRPV1-mediated mechanisms were also

involved after inflammation. Increasing anandamide levels for analgesic purposes by applying

substrate for its local synthesis may be superior to systemic anandamide application or

inhibition of its degradation.

Page 42 of 76



For Peer Review

3

Abbreviations

20:4-NAPE, N-arachidonoylphosphatidylethanolamine

AEA, anandamide, N-arachidonoylethanolamine

CB1, cannabinoid receptor 1

DRG, dorsal root ganglion

NAPE-PLD, N-acyl phosphatidylethanolamine phospholipase D

PWL, paw withdrawal latency

TRPV1, transient receptor potential cation channel subfamily V member 1

Tables of Links – instructions within the submission process

TARGETS

GPCRs

CB1 – ID 56

Voltage-gated ion channels

TRPV1 – ID 507

Enzymes

NAPE-PLD

(N-Acylphosphatidylethanolamine-phospholipase D)

LIGANDS

Anandamide – ID 2364

20:4-NAPE

PF514273, 2-(2-chlorophenyl)-3-(4-chlorophenyl)-7-(2,2-difluoropropyl)-6,7-dihydro-2H-

pyrazolo[3,4-f][1,4]oxazepin-8(5H)-one

SB366791 – ID 4309, (2E)-3-(4-chlorophenyl)-N-(3-methoxyphenyl)prop-2-enamide

Page 43 of 76



For Peer Review

4

Introduction

Modulation of synaptic transmission in the spinal dorsal horn is pivotal in nociceptive

signalling. An important part in this process is mediated by the transient receptor potential

cation channel, subfamily V, member 1 (TRPV1) and the Gi/o protein-coupled cannabinoid

type 1 (CB1) receptors (Katona and Freund, 2008; Nagy et al., 2014; Sousa-Valente et al.,

2014a). However, our current understanding does not allow us to utilise the putative analgesic

potential offered by controlling spinal signalling through these two receptors, fully.

Spinal TRPV1 is expressed predominantly by the central branches of nociceptive small- and

medium-sized dorsal root ganglion (DRG) neurons (Caterina et al., 1997). TRPV1 activation

has an excitatory effect through increasing the transmitter release from the terminals of DRG

neurons, but may depress evoked currents (Baccei et al. 2003).

CB1 receptor expression has been detected in various structures including inhibitory neurons,

astrocytes and central terminals of nociceptive small to medium size DRG neurons

(Ahluwalia et al., 2000; Alkaitis et al., 2010; Veress et al., 2013). Activation of both

presynaptic CB1 receptors and CB1 receptors on inhibitory interneurons leads to reduced

transmitter release from the respective neurons (Morisset and Urban, 2001; Pernia-Andrade et

al., 2009).

Several endogenous agents including N-arachidonoylethanolamine (anandamide; AEA)

activate both TRPV1 and the CB1 receptors (Ahluwalia et al., 2003; Devane et al., 1992;

Tognetto et al., 2001; Zygmunt et al., 1999). Importantly, sub-populations of DRG neurons as

well as spinal cord cells are able to synthesise or degrade anandamide (Carrier et al., 2004;

van der Stelt et al., 2005; Varga et al., 2014; Vellani et al., 2008).

Anandamide synthesis occurs through multiple metabolic pathways either in a Ca2+

-

insensitive or Ca2+

-sensitive manner (Ueda et al., 2013). N-

arachidonoylphosphatidylethanolamine (20:4-NAPE) constitutes the precursor for

anandamide synthesis in all pathways (Snider et al., 2010; Ueda et al., 2013; Wang and Ueda,

2009). We have shown recently that application of 20:4-NAPE to cultured DRG neurons

results in anandamide production in a concentration and temperature-dependent manner

(Varga et al., 2014).

Page 44 of 76



For Peer Review

5

Painful peripheral pathologies including tissue inflammation alter the expression and/or

activity of both TRPV1 and the CB1 receptor (Amaya et al., 2003; Amaya et al., 2006; Kanai

et al., 2005; Kwon et al., 2014; Luo et al., 2004; Richardson et al., 1998; Spicarova et al.,

2011; Spicarova and Palecek, 2009). Inflammation also induces changes in anandamide levels

in the spinal cord (Buczynski et al., 2010; Costa et al., 2010) and the expression of

anandamide synthesising and hydrolysing enzymes in DRG neurons (Lever et al., 2009;

Sousa-Valente et al., 2017). Here, for the first time, instead of “flooding” the entire

preparation by exogenous anandamide, we studied how providing substrate (20:4-NAPE) for

anandamide-synthesising pathways in the spinal cord affects nociceptive spinal synaptic

transmission and what role TRPV1 and CB1 receptors play in that process under naive and

inflammatory conditions.

Methods

Animals

All animal care and experimental procedures were in accordance with local Institutional

Animal Care and Use Committee and consistent with the guidelines of the International

Association for the Study of Pain, the U.K. Animals (Scientific Procedures) Act (1986) and

EU Directive 2010/63/EU for animal experiments. Animal studies are reported in compliance

with the ARRIVE guidelines (Kilkenny et al., 2010; McGrath and Lilley, 2015). Altogether

49 male Wistar rats (Institute of Physiology CAS, Czech Republic) of postnatal days (P19-

P23) were used in this study. Animals were maintained under temperature (22±2°C) and light-

controlled (12 h light/dark cycle) conditions with free access to food and water.

Spinal cord slice preparation

Male Wistar rats (P19-P23) were anaesthetised with isofurane (3%), the lumbar spinal cords

were removed and immersed in oxygenated ice-cold dissection solution containing (in mM)

95 NaCl, 1.8 KCl, 7 MgSO4, 0.5 CaCl2, 1.2 KH2PO4, 26 NaHCO3, 25 D-glucose and 50

sucrose. Animals were sacrificed by subsequent medulla interruption and exsanguination.

Each spinal cord was fixed to a vibratome stage (VT 1000S, Leica, Germany) using

cyanoacrylate glue in a groove between two agar blocks. Acute transverse slices 300-350 µm

thick were cut from L4-L5 segments, incubated in the dissection solution for 30min at 33°C,

stored in a recording solution at room temperature and allowed to recover for 1 h before the

electrophysiological experiments. Recording solution contained (in mM) 127 NaCl, 1.8 KCl,

1.2 KH2PO4, 2.4 CaCl2, 1.3 MgSO4, 26 NaHCO3, and 25 D-glucose. For the actual

Page 45 of 76



For Peer Review

6

measurement, slices were transferred into a recording chamber that was perfused continuously

with recording solution (room temperature) at a rate of ~2 ml.min-1

. All extracellular solutions

were saturated with carbogen (95% O2, 5% CO2) during the whole process.

Patch-clamp recording

Altogether, sEPSCs were recorded from 98 and eEPSCs from 79 superficial dorsal horn

neurons. Individual neurons were visualized using a differential interference contrast

microscope (DM LFSA, Leica, Germany) equipped with a 63×0.90 water-immersion

objective and an infrared-sensitive camera (KP-200P, Hitachi, Japan) with a standard

TV/video monitor (Hitachi VM-172, Japan). Patch pipettes were pulled from borosilicate

glass tubing when filled with intracellular solution; they had resistances of 3.5-6.0 MΩ. The

intracellular pipette solution contained (in mM) 125 gluconic acid lactone, 15 CsCl, 10

EGTA, 10 HEPES, 1 CaCl2, 2 MgATP, and 0.5 NaGTP and was adjusted to pH 7.2 with

CsOH. Voltage-clamp recordings in the whole cell configuration were performed with an

AxoPatch 200B amplifier (Molecular Devices, Sunnyvale, CA, USA) at room temperature

(21-24°C). Whole cell responses were low-pass filtered at 2 kHz and digitally sampled at 10

kHz. The series resistance of the recorded neurons was routinely compensated by 80% and

was monitored during the whole experiment. AMPA receptor-mediated spontaneous

excitatory postsynaptic currents (sEPSCs) were recorded from superficial dorsal horn neurons

in laminae I and II(outer), clamped at -70 mV in the presence of 10 µM bicuculline and 5 µM

strychnine. To evoke EPSCs, a dorsal root was stimulated with a suction electrode using a

constant current isolated stimulator (Digitimer, DS3, Hertfordshire, UK). Test pulses of

0.5 ms duration and an intensity ranging between 20 and 350 µA were applied at a frequency

of 0.033 Hz. The intensity of the stimulation was adjusted to evoke stable EPSCs. Application

of each drug lasted for 4 minutes period (recording solution, 20:4-NAPE, capsaicin, co-

application 20:4-NAPE and PF514273, co-application 20:4-NAPE and SB366791) or 6

minutes antagonist pretreatment (PF514273, SB366791, co-application of PF514273 with

SB366791). Neurons with capsaicin sensitive primary afferent input were identified by

increase of sEPSC frequency (>20%) following capsaicin (0.2 µM) application at the end of

the experimental protocol. Capsaicin was applied in 87% of the recorded neurons and 92% of

these responded with sEPSC frequency increase. The software package pCLAMP version

10.0 (Axon Instruments, CA, USA) was used for data acquisition and subsequent off-line

analysis. Cells with a series resistance >20 MΩ were excluded from the analysis.

Page 46 of 76



For Peer Review

7

Anandamide release experiments

Spinal cord slices from 5 animals were prepared in the same way as for the

electrophysiological experiments. For each of the 5 experiments, 18 acute spinal cord slices

were used. Slices were put into plastic safe-lock tube with 200 µl of the recording solution,

saturated by carbogen (95% O2 - 5% CO2) during the whole process. Incubation of 10 min

was used before the whole volume of the solution (sample) was extracted and immediately

frozen for later mass spectrometry analysis. The solution in the tube was immediately

replaced with another solution sample. In each experiment 8 samples were taken after 10 min

of incubation: 1 (recording solution), 2 (recording solution), 3 (20:4-NAPE, 20 µM), 4

(recording solution), 5 (20:4-NAPE, 100 µM), 6 (recording solution), 7 (20:4-NAPE, 200

µM), 8 (recording solution). Additional samples with only 20:4-NAPE, 20-100-200 µM

without the slices were also prepared and analysed.

The collected samples were analysed for the presence of AEA with mass spectrometry. For

calibration purposes solutions of AEA and 20:4-NAPE were used. Anandamide content was

determined by reversed-phase high-performance chromatography using Agilent 1100 LC

system (Agilent, Palo Alto, CA, USA) consisting of a degasser, a binary pump, an

autosampler, and an thermostatic column compartment. Chromatographic separation was

carried out in a Kinetex 2.6u PFP, 100A column (100 x 2.1 mm I.D., Phenomenex, Torrence,

CA, USA). The sample (10 µl) was injected into the column and eluted with a gradient

consisting of (A) water-formic acid 100:0.1 v/v and (B) acetontrile-formic acid 100:0.085 v/v

(flow rate 0.35 ml.min-1

and temperature 40°C). The gradient started at A/B 80:20 for 5 min,

reaching 100% B after 10 min. For the next 5 min the elution was isocratic at 100% B.

Elution was monitored by an ion-trap mass spectrometer (Agilent LC-MSD Trap XCT-Ultra;

Agilent, Palo Alto, CA, USA). Atmospheric pressure ionization-electrospray ionization (API-

ESI) positive mode ion-trap mass spectrometry at MRM (multiple reaction monitoring) mode

was used with transition of m/z 348.1>287.1 for anandamide (retention time 11.2 min) when

monitored mass range was 100-400 m/z. Operating conditions: drying gas (N2), 12 l.min-1

;

drying gas temperature, 350°C; nebulizer pressure, 30 psi (207 kPa). The areas of the

anandamide peak were measured. The results from each experiment (peak areas of AEA)

were normalized to the AEA production after the 200 µM 20:4-NAPE application (100%).

Peripheral inflammation

Peripheral inflammation was induced in a group of animals 24 h before the spinal cord slice

preparation was made. Under isoflurane (3%) anaesthesia, both hind paws were injected

Page 47 of 76



For Peer Review

8

subcutaneously by a 3% mixture of carrageenan (50 µl) in a physiological saline solution. The

animals were left to recover in their home cages. This carrageenan injection in peripheral

tissue is thoroughly characterized and established animal model of inflammatory pain (Ren

and Dubner, 1999). Naive animals were used as controls.

Behavioural testing

The animals used in model of peripheral inflammation were tested for responsiveness to

thermal stimuli before and 24 h after the model induction. Paw withdrawal latencies (PWLs)

to radiant heat stimuli were determined for both hind paws using Plantar Test 37370 apparatus

(Ugo Basile, Italy). The rats were placed under non-binding, clear plastic cages on a 3 mm

thick glass plate and left to adapt at least for 20min. The radiant heat was applied to the

plantar surface of each hind paw until a deliberate escape movement of the paw was detected

by Plantar Test apparatus. The PWLs were tested 4 times for each hind paw with at least 5min

intervals between the trials. Results from each hind paw were averaged. Baseline withdrawal

latencies were determined in all animals before any experimental procedure.

Data and statistical analysis

The data and statistical analysis comply with the recommendations on experimental design

and analysis in pharmacology (Curtis et al., 2015). Some data were normalized as a

percentage of the control values (100%). Results are shown as means ± SEM. For offline

analysis of the recorded sEPSCs data, segments of 2-min duration were used for each

experimental condition. Only sEPSCs with amplitudes 5 pA or greater (which corresponded

to at least twice the noise level) were included in the frequency analysis. In the case of

amplitude analysis, the same sEPSCs events were used. Statistics were calculated using

SigmaStat 3.5 (Systat Software, CA, USA). A Kolmogorov-Smirnov test was used to evaluate

statistical significance for cumulative data. One-way ANOVA or one-way ANOVA repeated

measures (RM) followed by a post hoc test (Student-Newman-Keuls) or paired t-test was used

for data with normal distribution and non-parametric rank test or RM on ranks was used

where appropriate for statistical comparisons. P-value <0.05 was considered statistically

significant. Detailed information is given in the figure legends.

Materials

All chemicals used for extracellular and intracellular solutions were of analytical grade and

purchased from Sigma Aldrich (St. Louis, MO, USA) and Tocris Bioscience (Bristol, UK).

Page 48 of 76



For Peer Review

9

Capsaicin, SB366791 and PF514273 (Tocris Bioscience) and anandamide (Avanti Polar

Lipids, Alabaster, AL, USA) were dissolved in dimethylsulfoxide (DMSO; Sigma Aldrich),

which had a concentration <0.1% in the final solution. 20:4-NAPE (Avanti Polar Lipids) was

dissolved in chloroform, which had a concentration <0.1% in the final solution. Concentration

of selective TRPV1 antagonist SB366791 (10 µM) was based on our previous studies

(Spicarova et al., 2014a; Spicarova and Palecek, 2009) and the selectivity pA2 = 7.71

(Gunthorpe et al., 2004). Concentration of highly selective CB1 antagonist PF514273 (0.2

µM) was determined by considering Ki - 1 nM (Dow et al., 2009) and the needed diffusion

through the spinal cord slice. 20:4-NAPE was applied in the recording solution in

concentration (20 µM) based on our preliminary results and previous experiments performed

on DRG cultures (Varga et al., 2014). Carrageenan for induction of inflammation was

purchased from Sigma Aldrich.

Results

Application of 20:4-NAPE increased anandamide concentration in spinal cord slices

To verify the production of anandamide from 20:4-NAPE in our preparation, mass

spectrometry was used to analyse AEA content after application of different concentrations of

20:4-NAPE (20 µM, 100 µM and 200 µM) on spinal cord slices in vitro. Under the control

conditions with extracellular solution only, the average AEA concentration in the solution was

very low (7067±4532 of peak area, n=5, Fig. 1). AEA concentration increased gradually with

increasing concentration of 20:4-NAPE application (20 µM: 48324±27502; 100 µM:

103310±38179; 200 µM: 298004±139867 AEA peak area, n=5). To reduce the differences

between the individual experiments, the results were standardized for the statistical analysis

(Fig. 1). There was no AEA detected in the samples where 20:4-NAPE was present without

the slices. These results indicate that 20:4-NAPE (20 µM) application in our

electrophysiological experiments led to increased AEA concentration in the spinal cord slice.

Application of 20:4-NAPE reduced both spontaneous and evoked activity of spinal

dorsal horn neurons

The role of 20:4-NAPE in nociceptive synaptic transmission was investigated using

recordings of spontaneous EPSC and dorsal root stimulation-evoked EPSCs from neurons in

laminae I and II(outer) of the dorsal horn. Altogether the mean control frequency of sEPSCs

recorded in neurons from slices prepared from naive animals was 1.09±0.14 Hz (n=42).

Page 49 of 76



For Peer Review

10

Application of 20:4-NAPE (20 µM, 4min) robustly decreased sEPSC frequency in 12 of the

13 recorded neurons to 52.3±7.8% (n=13; Fig. 2A,B,C) when data was averaged from all

neurons in this group. The average amplitude of the sEPSCs was also reduced significantly

after the 20:4-NAPE application from 24.9±2.5 pA to 21.4±1.4 pA (n=13, p<0.05, paired t-

test). However, the decrease of the amplitude (>15%) was present in only 4 cells out of 13

neurons and the cumulative distribution of sEPSC amplitudes did not show significant change

after the 20:4-NAPE application (Fig. 2D).

Similarly to the changes in sEPSC’s, 20:4-NAPE (20 µM, 4min) also significantly decreased

the amplitude of eEPSCs to 70.5±9.0% (n=15; Fig. 2E,F). The reduction of eEPSCs

amplitude (>15%) was present in 8 of 15 recorded neurons. Together, these findings indicate

that application of 20:4-NAPE had a robust inhibitory effect on the excitation of superficial

spinal dorsal horn neurons in naive conditions.

The 20:4-NAPE-induced inhibitory effect on spontaneous activity was prevented by

blocking the CB1 but not TRPV1 receptors

In the next experiments we have investigated whether the 20:4-NAPE-induced inhibitory

effect is mediated through either of the anandamide’s main targets, the CB1 and TRPV1

receptors.

Application of the highly selective CB1 antagonist PF514273 (0.2 µM, 6min) caused a small

numerical increase of the sEPSCs frequency; however this enhancement did not reach

statistical significance (112.6±14.8%, n=11; Fig. 3A,C). Subsequent co-application of

PF514273 (0.2 µM) and 20:4-NAPE (20 µM, 4min) did not change the frequency of sEPSCs

compared to the control value and to the antagonist pretreatment (98.1±13.4%; Fig. 3A,C,F).

The amplitude of sEPSCs in control conditions (20.0±1.7 pA, n=11) was not affected either

by pre-application alone (18.8±1.8 pA) or the subsequent co-application of PF514273 and

20:4-NAPE (18.6±2.1 pA). The failure of 20:4-NAPE reducing either the frequency or the

amplitudes of sEPSCs in the presence of PF514273 is in contrast with the effect of 20:4-

NAPE alone (Fig. 2B).

Pretreatment with the selective TRPV1 antagonist SB366791 (10 µM, 6min) alone did not

change the sEPSC frequency (97.6±13.9%, n=10; Fig. 3B,D). However, co-application of

20:4-NAPE (20 µM, 4min) and SB366791 (10 µM) significantly reduced the frequency of

Page 50 of 76



For Peer Review

11

sEPSCs in 8 of 10 recorded neurons (53.0±14.7%, n=10; Fig. 3B, D) when compared to the

control values. The degree of reduction was not different from that 20:4-NAPE produced

alone (Fig. 2B). The amplitude of sEPSCs was not affected by the SB366791 pretreatment

(control: 29.0±4.1 pA, SB366791: 24.6±4.0 pA) or following SB366791 with 20:4-NAPE

treatment (25.2±4.6 pA, n=10).

Further, combined pretreatment with both CB1 and TRPV1 antagonists PF514273 (0.2 µM)

and SB366791 (10 µM, 6min) was tested. This pre-treatment caused a numerical increase of

the sEPSCs frequency that was not statistically significant (122.0±11.4%, n=8; Fig. 3E).

Subsequent co-application of both antagonists PF514273 (0.2 µM) and SB366791 (10 µM)

together with 20:4-NAPE (20 µM, 4min) did not change the frequency of sEPSCs compared

to the control value (100.2±7.6%; Fig. 3E). The average amplitude of sEPSCs was not

changed during entire experiment (control: 20.8±2.0 pA, PF514273 + SB366791: 19.2±1.6

pA, PF514273 + SB366791 + 20:4 NAPE: 17.4±1.1 pA, n=8).

To compare all the experimental situations and to diminish any influences of the antagonists’

applications alone, we have also analysed the data in a way where the antagonist application

together with 20:4-NAPE was expressed as a percentage of the previous condition (Fig. 3F).

PF514237 + 20:4-NAPE compared to the PF514273 pretreatment (87.5±5.3%, n=11);

SB366791 + 20:4-NAPE compared to the SB366791 pretreatment (58.1±16.0%, n=10) and

PF514237 + SB366791 + 20:4-NAPE compared to both antagonists pretreatment (84.6±6.6%,

n=8). This analysis confirmed the overall differential effect of the antagonists. These findings

taken together show that the 20:4-NAPE application-induced inhibitory effect on the

frequency of sEPSCs is mediated by activation of CB1 receptors, but not by TRPV1

receptors.

20:4-NAPE-induced reduction of eEPSCs amplitude was prevented by CB1 but not

TRPV1 antagonist in naive animals.

The respective antagonists of the CB1 and TRPV1 receptors PF514273 and SB366791 were

used to identify the contribution of these receptors to the 20:4-NAPE-induced decrease of

eEPSC amplitude. Application of PF514273 (0.2 µM, 6min) did not significantly change the

amplitude of eEPSCs (89.0±6.6%, n=13; Fig. 4A,C). Subsequent co-application of PF514273

(0.2 µM, 4min) and 20:4-NAPE (20 µM) did not change significantly the average amplitude

of the eEPSC compared to both the control (77.0±10.5%, n=13; Fig. 4C) and PF514273

Page 51 of 76



For Peer Review

12

pretreatment (88.5±10.1%; Fig. 4E) values. Out of these neurons 7 exhibited a lack of

reduction. Four of these 7 recorded neurons were not affected by 20:4-NAPE application and

in remaining 3 neurons the amplitude increased >15%. These results show that the inhibitory

effect of 20:4-NAPE on the eEPSCs amplitudes is mediated by CB1 receptors in the group of

superficial spinal dorsal horn neurons.

Inhibition of TRPV1 receptors by SB366791 (10 µM, 6min) alone did not change the eEPSC

amplitude (108.8±14.7%, n=10; Fig. 4B,D). Subsequent co-application of SB366791 (10 µM,

4min) and 20:4-NAPE (20 µM) did not prevent the amplitude decrease induced by the 20:4-

NAPE, compared to both control (64.9±9.7%, n=10; Fig. 4D) and SB366791 pretreatment

(64.2±9.3%; Fig. 4E) values. The amplitude reduction was evident in 8 of 10 neurons and it

did not change in the 2 remaining neurons. The degree of 20:4-NAPE-induced reduction in

the presence of SB366791 was not significantly different from that produced by 20:4-NAPE

alone (Fig. 4E). Hence, these findings indicate that TRPV1 is not involved in mediating the

20:4-NAPE-induced inhibitory effect on eEPSC amplitude.

Application of 20:4-NAPE reduced the frequency of sEPSCs in spinal dorsal horn

neurons under inflammatory conditions

Peripheral inflammation was induced by subcutaneous injection of carrageenan 24 hours

before behavioural testing. Signs of inflammation (redness, hypersensitivity and swelling)

were present at the hind paws of all animals. The paw withdrawal latency to thermal stimuli

was significantly decreased from 11.82±0.60 s to 8.34±0.51 s (n=12, p<0.05, paired t-test).

The control sEPSC frequency 1.28±0.24 Hz (n=56) recorded in neurons 1 day after the

inflammation induction was higher but not statistically different when compared to the control

sEPSC frequency recorded in naive animals.

Application of 20:4-NAPE (20 µM, 4min) to slices prepared from the spinal cord of these

animals strongly inhibited the sEPSC frequency in 7 of the 9 recorded neurons (59.5±15.6%,

n=9; Fig. 5). This 20:4-NAPE-induced inhibitory effect on frequency of sEPSC under

inflammatory conditions was not significantly different from that observed in naive rats

(52.3±7.8%, n=13; Fig. 2B). Application of 20:4-NAPE also reduced the amplitude of

sEPSCs from 21.4±2.3 pA to 18.4±1.6 pA, (n=9, p=0.05, paired t-test). However, decrease of

more than 15% was present only in 3 from 9 recorded neurons.

Page 52 of 76



For Peer Review

13

The inhibitory effect of 20:4-NAPE on sEPSCs was mediated through the CB1 with

TRPV1 receptors participation under inflammatory conditions

Blocking the CB1 and TRPV1 receptors was tested on the 20:4-NAPE-induced inhibitory

effect in slices after the induction of peripheral inflammation. The CB1 receptor antagonist

PF514273 (0.2 µM, 6min) pretreatment did not induce any change in the sEPSC frequency

(108.0±9.8%, n=16; Fig. 6A,C). Subsequent co-application of PF514237 (0.2 µM) and 20:4-

NAPE (20 µM) did not change significantly the sEPSCs frequency when compared to the

control value (148.8±16.8%, n=16; Fig. 6A,C). In 9 of these 16 recorded neurons the

frequency of sEPSC increased, in 4 neurons it did not change and it decreased in 3 neurons.

The amplitude of sEPSCs was not significantly affected by PF514273 application alone

(control: 27.4±2.4 pA, PF514273: 24.6±2.5 pA) or PF514273 with 20:4-NAPE co-application

(24.1±2.2 pA, n=16).

The application of the TRPV1 antagonist SB366791 (10 µM, 6min) significantly decreased

the frequency of sEPSCs (71.5±10.9%, n=16; Fig. 6B,D). Subsequent co-application of

SB366791 (10 µM) and 20:4-NAPE (20 µM) induced a further decrease of the sEPSC

frequency compared to the control value (55.2±12.9%; Fig. 6B,D). When responses of the

individual cells were assessed, 11 of the 16 neurons exhibited decrease in sEPSC frequency

and it did not change in the rest of the cells. The amplitude of sEPSCs was not significantly

changed during the entire experiment (control: 24.2±1.7 pA, SB366791: 24.2±1.6 pA,

SB366791 + 20:4-NAPE: 21.8±1.2 pA, n=16).

The effect of combined inhibition of both receptors was evaluated in further experiments.

Pretreatment with PF514273 (0.2 µM) and SB366791 (10 µM) did not significantly change

the sEPSC frequency when all the neurons were pooled together (99.6±12.7%, n=15; Fig.

6E). Although in 9 of these 15 recorded neurons antagonists co-application decreased the

sEPSC frequency (64.9±4.8%, p<0.05, RM ANOVA on ranks followed by Student-Newman-

Keuls test), in 5 neurons it increased the sEPSC frequency (160.6±11.2%, p>0.05, RM

ANOVA on ranks). Subsequent co-application of PF514273 (0.2 µM), SB366791 (10 µM)

and 20:4-NAPE (20 µM) decreased the sEPSC frequency (76.9±11.6%, n=15; Fig. 6E)

compared to the control values. Antagonist co-treatment prevented the inhibitory effect of

20:4-NAPE application in 7 of the 15 cells. The amplitude of sEPSCs was not significantly

Page 53 of 76



For Peer Review

14

changed during different recording conditions (control: 24.9±3.7 pA, SB366791 + PF514273:

23.9±3.2 pA, SB366791 + PF514273 + 20:4-NAPE: 20.0±2.8 pA, n=15).

In addition, in order to compare the overall effect of the 20:4-NAPE application under the

different experimental conditions, the data were analysed also as a percentage of the previous

condition (=100%) and the differences statistically evaluated (Fig. 6F). Under this assessment

the sEPSC frequency after the 20:4-NAPE application alone was 59.5±15.6% of the control

(n=9). The sEPSC frequency after PF514237 + 20:4-NAPE application as a percentage of the

PF514273 pretreatment was significantly increased (152.7±21.4%, n=16), the sEPSC

frequency during the SB366791 + 20:4-NAPE application as a percentage of SB366791

pretreatment was (77.8±10.6%, n=16) and PF514237 + SB366791 + 20:4-NAPE as a

percentage of the PF514237 + SB366791 pretreatment was 81.2±8.2% (n=15). The increase

after the PF514237 + 20:4-NAPE application was significantly different from all the other

conditions.

These results suggest that the inhibitory effect induced by 20:4-NAPE application on the

sEPSC frequency is preferentially mediated by activation of CB1 receptors under the

inflammatory conditions (Fig. 6F). Moreover, when CB1 receptors were blocked the 20:4-

NAPE application led to increased sEPSC frequency, that was prevented by TRPV1 receptor

inhibition.

The reduction of the eEPSC amplitude induced by application of 20:4-NAPE was

prevented by blocking either the CB1 or TRPV1 receptors under the inflammatory

conditions

The 20:4-NAPE (20 µM) application during recording of eEPSCs in dorsal horn neurons after

dorsal root stimulation in spinal cord slices prepared 24 h after induction of peripheral

inflammation, significantly decreased the eEPSCs amplitude (78.5±6.6%, n=14; Fig. 7). This

decrease (>15%) was present in 9 of the 14 recorded neurons.

Treatment of the slices with CB1 receptor antagonist PF514273 (0.2 µM, 6min) did not

change the eEPSCs amplitude (109.0±13.6%, n=16, Fig. 8A,C). Subsequent co-application of

PF514273 (0.2 µM, 4min) and 20:4-NAPE (20 µM) increased the amplitude of eEPSCs

without reaching statistical significance (125.1±26.6%, Fig. 8A,C) when compared to the

Page 54 of 76



For Peer Review

15

control values. A CB1 receptor antagonist thus prevented the inhibitory effect induced by

application of 20:4-NAPE in 11 of 16 neurons.

Application of the TRPV1 antagonist SB366791 (10 µM, 6min) in another group of neurons

did not change the eEPSC amplitude either (88.0±10.8%, n=11, Fig. 8B,D). Subsequent co-

application of SB366791 (10 µM, 4min) and 20:4-NAPE (20 µM) did not change the

amplitude of eEPSC (92.0±17.8%, Fig. 8B,D) compared to the control values. Similarly to the

inhibition of CB1 receptors, the TRPV1 receptor antagonist prevented the 20:4-NAPE-

induced inhibitory effect in majority (8 of 11) of the recorded superficial dorsal horn neurons.

In addition, to compare the 20:4-NAPE effect between the different experimental conditions,

the same data were expressed as a percentage of the previous application: 20:4-NAPE as

percentage of eEPSC basal amplitude (78.5±6.6%, n=14); PF514237 + 20:4-NAPE as a

percentage of PF514273 pretreatment (117.7±17.7%, n=16); SB366791 + 20:4-NAPE as a

percentage of SB366791 pretreatment (101.3±16.0%, n=11, Fig. 8E). These results indicate

that under the inflammatory conditions both CB1 and TRPV1 receptors mediated the

inhibitory effect induced by the 20:4-NAPE application on evoked EPSC amplitude.

Discussion

Here we report that 20:4-NAPE application induced an inhibitory effect on excitatory

nociceptive synaptic transmission demonstrated by decrease of sEPSC frequency and

reduction of dorsal root stimulation-evoked EPSC amplitude in the superficial spinal dorsal

horn. The inhibitory effect occurred both in naive conditions and following the development

of hindpaw inflammation. The differential effects of CB1 and TRPV1 antagonists indicated

that the underlying mechanisms of 20:4-NAPE-induced inhibition may differ in those two

conditions.

20:4-NAPE and anandamide synthesis

20:4-NAPE is a substrate for anandamide synthesis in enzyme preparations and cultured

primary sensory neurons (Varga et al., 2014; Wang et al., 2006). Here we found that spinal

cord slices also produce AEA after 20:4-NAPE application. Although direct effects of 20:4-

NAPE, or indirect effects through a metabolite other than anandamide on some other

Page 55 of 76



For Peer Review

16

receptors cannot be categorically excluded, we propose that at least the great majority of the

20:4-NAPE application-induced effects described here, was mediated through the synthesis of

anandamide acting on CB1 and TRPV1 receptors.

Anandamide activation of CB1 and TRPV1 receptors

The CB1 receptor and TRPV1 constitute the main targets for anandamide (Devane et al.,

1992; Zygmunt et al., 1999) and our data show that those two receptors mediate at least the

majority of the 20:4-NAPE application-induced effects. However, as anandamide is a highly

promiscuous molecule, the involvement of other molecules including peroxisome proliferator-

activated receptor alpha and gamma, sodium and T-type Ca2+

channels (Kim et al., 2005;

Okura et al., 2014; Chemin et al., 2001; O'Sullivan, 2007) cannot be ruled out. Nevertheless,

due to the robust effects of the CB1 receptor and TRPV1 antagonists, here we addressed the

contribution of these two receptors only.

The 20:4-NAPE application-induced inhibitory effect on sEPSC frequency and eEPSC

amplitude was mediated preferentially by activation of CB1 receptor under naive conditions.

Although post-synaptic CB1 receptor expression in the spinal cord has been reported

(Farquhar-Smith et al., 2000), most studies suggest exclusive pre-synaptic location either on

DRG neuron terminals or terminals of GABAergic inhibitory interneurons (Hegyi et al.,

2012; Nyilas et al., 2009; Pernia-Andrade et al., 2009; Veress et al., 2013). CB1 receptor

activation at both locations leads to reduced transmitter release (Morisset et al., 2001; Nyilas

et al., 2009; Pernia-Andrade et al., 2009). In our preparations the inhibitory synaptic

transmission was pharmacologically blocked. Therefore, it seems plausible to suggest that the

CB1 receptor-mediated inhibitory effect by 20:4-NAPE application occurred through

anandamide-mediated CB1 receptor activation and subsequent reduction of transmitter release

from spinal terminals of DRG neurons.

Under the naive condition, there was a tendency of CB1 receptors antagonist per se to mildly

increase the frequency of sEPSC, although this increase did not reach statistical significance.

TRPV1 antagonist did not have any effect on the superficial dorsal horn neurons sEPSC

frequency, similar to our previous experiments (Spicarova et al., 2014a). Nevertheless

moderate TRPV1 mediated sEPSC tonic activity was reported in lamina II neurons in mice

(Park et al., 2011).

Page 56 of 76



For Peer Review

17

The robust decrease of AMPA receptor-mediated sEPSC frequency (Spicarova and Palecek,

2010) induced by 20:4-NAPE application was also accompanied by moderate reduction of

sEPSC amplitude in some neurons. In our preparation the recorded neurons have contacts

with numerous synapses, which spontaneously release glutamate and induce sEPSCs. The

robust decrease of sEPSC frequency could elicit strong attenuation of glutamate release from

specific, 20:4-NAPE application-responding afferents leading to average sEPSC amplitude

decrease without affecting the postsynaptic mechanisms.

The effect of peripheral inflammation

In slices taken after the induction of peripheral inflammation, 20:4-NAPE application induced

a significant inhibition of sEPSC frequency similar to the naive preparations. However,

SB366791 reduced sEPSC frequency, suggesting presynaptic TRPV1 receptors tonic

activation. The effect of SB366971 per se is consistent with inflammation-induced tonic

activity (Lappin et al., 2006) and increased sensitivity to endogenous agonists (Spicarova and

Palecek, 2009) of presynaptic TRPV1 in the spinal cord dorsal horn. TRPV1 is expressed in

the overwhelming majority of spinal C-fibre terminals in the superficial dorsal horn (Caterina

et al., 1997; Guo et al., 1999). Consistently with this high TRPV1 expression, regulation

(activation, desensitisation and inhibition) of TRPV1 has large impact on glutamate release

from these afferents (Spicarova et al., 2014b). It was suggested that modulation of TRPV1 in

the dorsal horn could underlie several pathological pain states (Kanai et al., 2005; Spicarova

et al., 2014a; Spicarova et al., 2011).

Tonic activation of presynaptic CB1 receptors was not detected under the inflammatory

conditions. However, the CB1 receptor antagonist prevented the 20:4-NAPE application-

produced inhibitory effect on sEPSC frequency. Moreover, 20:4-NAPE application

significantly increased the frequency of sEPSCs, when CB1 receptors were blocked and this

potentiating effect was prevented by TRPV1 receptor inhibition (Fig. 6F). This indicates that

under inflammatory conditions 20:4-NAPE application-induced inhibition of the sEPSC

frequency was mediated by CB1 receptors while the potentiating effect mediated by TRPV1

receptors was unmasked only when CB1 receptors were blocked.

The CB1 receptor-mediated block of 20:4-NAPE application-induced inhibitory effect on

eEPSC amplitude was maintained after the development of inflammation. However, this 20:4-

NAPE application-induced inhibitory effect was prevented by blocking either CB1 or TRPV1

Page 57 of 76



For Peer Review

18

receptors, indicating involvement of both receptors. We did not observe a significant

reduction of eEPSC amplitude after TRPV1 antagonist application as with the sEPSC. While

it is possible that activation of TRPV1 receptors under these conditions did not play such an

important role, it needs also to be taken into account that the electrical stimulation of dorsal

root could activate also myelinated primary afferents that do not express TRPV1 receptors

(Caterina et al., 1997; Guo et al., 1999). The effect of the TRPV1 antagonist application thus

could be “diluted”.

In contrast to potentiation of the spontaneous transmitter release by TRPV1 agonist, the

release induced by action potentials evoked by dorsal root electrical stimulation may be

blocked by TRPV1 receptors activation (Yang et al., 1999, Baccei et al., 2003). Thus it is

plausible, that activation of TRPV1 on presynaptic terminals of DRG neurons by 20:4-NAPE

application reduced the glutamate release from primary afferents and thus contributed to the

decrease of evoked EPSC amplitude in the recorded postsynaptic neuron. In addition, rapid

internalization of voltage activated Ca2+

channels by TRPV1 activation (Wu et al., 2005)

could underlie the reduction of synchronous transmitter release. Although, the vast majority

of spinal TRPV1 is localized in terminals of primary sensory neurons, postsynaptic TRPV1

expression was also described in some GABAergic neurons, in which TRPV1 activation

induces long-term depression through the reduction of AMPA channels in the plasmatic

membrane (Caterina et al., 1997; Guo et al., 1999; Kim et al., 2012). We cannot exclude the

possibility that our neurons recorded in laminae I and II(outer) could include GABAergic cells

in which the postsynaptic TRPV1-mediated modulation under the inflammatory conditions

could occur, though it would change only the EPSC amplitude.

The role of 20:4-NAPE and anandamide in nociceptive modulation

In summary, our data together indicate that 20:4-NAPE application induces mainly CB1

receptor mediated inhibitory effects on excitatory transmission in naive animals while TRPV1

mediated mechanisms are also involved after peripheral inflammation. We propose, that if the

20:4-NAPE application-induced effects are indeed mediated through anandamide synthesis,

balanced signalling by anandamide and its targets are involved in preventing the spread of

nociceptive signals into supraspinal structures and this balance may be compromised during

inflammation.

Page 58 of 76



For Peer Review

19

Anandamide, due to its lipophilic nature, is expected to be produced in close proximity to its

target. TRPV1-expressing primary sensory neurons indeed express multiple anandamide-

synthesising pathways (Sousa-Valente et al., 2014b; Sousa-Valente et al., 2017; Varga et al.,

2014). Further, transcripts of several anandamide-synthesising enzymes are expressed in the

spinal dorsal horn (Malek et al., 2014). Enhanced activity after inflammation and during the

electrical stimulation of primary afferent fibres resulted in increased concentration of Ca2+

in

presynaptic terminals and could induce or increase the enzymatic activity of NAPE-PLD.

Furthermore Ca2+

influx through postsynaptic AMPA receptors could be also involved

through promoting anandamide synthesis from 20:4-NAPE, as NAPE-PLD is expressed in

post-synaptic dendrites in the spinal dorsal horn (Hegyi et al., 2012). Ca2+

-insensitive

pathways and NAPE-PLD activity as a part of a retrograde inhibitory mechanism (Katona and

Freund, 2008) could be also involved in this anandamide synthesis. NAPE-PLD and other

anandamide-synthesising enzymes may be a particularly important for regulating nociceptive

spinal processing under inflammatory conditions.

The 20:4-NAPE application in our experiments also provided a distinctive opportunity to

study the role of the spinal endocannabinoid system, by application of substrate for

anandamide synthesis instead of anandamide directly. By this approach, physiological

mechanisms of anandamide synthesis played an important role, including the level of their

activity and local distribution, creating anandamide microdomains concentrations. By

flooding the preparation by anandamide directly, most likely other receptors and biological

pathways would have been activated. This method of local “on demand” anandamide

production from its precursor may prove to be of advantage also in the clinical settings for

pain treatment. Especially, as clinical trials focused to increase anandamide levels by reduced

hydrolysis with fatty acid amide hydrolase inhibitors did not show clinical efficacy (Mallet et

al., 2016).

Page 59 of 76



For Peer Review

20

Acknowledgements

This work was supported by GACR 15-11138S, MSMT LH15279, CZ.1.05/1.1.00/02.0109,

RVO67985823, GAUK138215. Authors would like to thank Prof. Ivan Miksik for the AEA

mass spectrometry analysis.

Author contributions

JP conceived and designed the study. VN, PM and PA conducted experiments, VN, PM and

DS analysed the data. VN, IN, DS and JP participated in writing the manuscript. All authors

read and approved the final version of the manuscript.

Conflict of interest

The authors declare no conflicts of interest.

Declaration of transparency and scientific rigour

This Declaration acknowledges that this paper adheres to the principles for transparent

reporting and scientific rigour of preclinical research recommended by funding agencies,

publishers and other organisations engaged with supporting research.

Page 60 of 76



For Peer Review

21

References

Ahluwalia J, Urban L, Bevan S, Nagy I (2003). Anandamide regulates neuropeptide release from

capsaicin-sensitive primary sensory neurons by activating both the cannabinoid 1 receptor and the

vanilloid receptor 1 in vitro. The European journal of neuroscience 17(12): 2611-2618.

Ahluwalia J, Urban L, Capogna M, Bevan S, Nagy I (2000). Cannabinoid 1 receptors are expressed in

nociceptive primary sensory neurons. Neuroscience 100(4): 685-688.

Alkaitis MS, Solorzano C, Landry RP, Piomelli D, DeLeo JA, Romero-Sandoval EA (2010). Evidence for a

role of endocannabinoids, astrocytes and p38 phosphorylation in the resolution of postoperative

pain. PloS one 5(5): e10891.

Amaya F, Oh-hashi K, Naruse Y, Iijima N, Ueda M, Shimosato G, et al. (2003). Local inflammation

increases vanilloid receptor 1 expression within distinct subgroups of DRG neurons. Brain research

963(1-2): 190-196.

Amaya F, Shimosato G, Kawasaki Y, Hashimoto S, Tanaka Y, Ji RR, et al. (2006). Induction of CB1

cannabinoid receptor by inflammation in primary afferent neurons facilitates antihyperalgesic effect

of peripheral CB1 agonist. Pain 124(1-2): 175-183.

Baccei ML, Bardoni R, Fitzgerald M (2003). Development of nociceptive synaptic inputs to the

neonatal rat dorsal horn: glutamate release by capsaicin and menthol. The Journal of physiology

549(Pt 1): 231-242.

Buczynski MW, Svensson CI, Dumlao DS, Fitzsimmons BL, Shim JH, Scherbart TJ, et al. (2010).

Inflammatory hyperalgesia induces essential bioactive lipid production in the spinal cord. Journal of

neurochemistry 114(4): 981-993.

Carrier EJ, Kearn CS, Barkmeier AJ, Breese NM, Yang W, Nithipatikom K, et al. (2004). Cultured rat

microglial cells synthesize the endocannabinoid 2-arachidonylglycerol, which increases proliferation

via a CB2 receptor-dependent mechanism. Molecular pharmacology 65(4): 999-1007.

Caterina MJ, Schumacher MA, Tominaga M, Rosen TA, Levine JD, Julius D (1997). The capsaicin

receptor: a heat-activated ion channel in the pain pathway. Nature 389(6653): 816-824.

Chemin J, Monteil A, Perez-Reyes E, Nargeot J, Lory P (2001). Direct inhibition of T-type calcium

channels by the endogenous cannabinoid anandamide. The EMBO journal 20(24): 7033-7040.

Costa B, Bettoni I, Petrosino S, Comelli F, Giagnoni G, Di Marzo V (2010). The dual fatty acid amide

hydrolase/TRPV1 blocker, N-arachidonoyl-serotonin, relieves carrageenan-induced inflammation and

hyperalgesia in mice. Pharmacological research : the official journal of the Italian Pharmacological

Society 61(6): 537-546.

Curtis MJ, Bond RA, Spina D, Ahluwalia A, Alexander SP, Giembycz MA, et al. (2015). Experimental

design and analysis and their reporting: new guidance for publication in BJP. British journal of

pharmacology 172(14): 3461-3471.

Devane WA, Hanus L, Breuer A, Pertwee RG, Stevenson LA, Griffin G, et al. (1992). Isolation and

structure of a brain constituent that binds to the cannabinoid receptor. Science 258(5090): 1946-

1949.

Dow RL, Carpino PA, Hadcock JR, Black SC, Iredale PA, DaSilva-Jardine P, et al. (2009). Discovery of 2-

(2-chlorophenyl)-3-(4-chlorophenyl)-7-(2,2-difluoropropyl)-6,7-dihydro-2H-pyraz olo[3,4-

Page 61 of 76



For Peer Review

22

f][1,4]oxazepin-8(5H)-one (PF-514273), a novel, bicyclic lactam-based cannabinoid-1 receptor

antagonist for the treatment of obesity. Journal of medicinal chemistry 52(9): 2652-2655.

Farquhar-Smith WP, Egertova M, Bradbury EJ, McMahon SB, Rice AS, Elphick MR (2000). Cannabinoid

CB(1) receptor expression in rat spinal cord. Molecular and cellular neurosciences 15(6): 510-521.

Gunthorpe MJ, Rami HK, Jerman JC, Smart D, Gill CH, Soffin EM, et al. (2004). Identification and

characterisation of SB-366791, a potent and selective vanilloid receptor (VR1/TRPV1) antagonist.

Neuropharmacology 46(1): 133-149.

Guo A, Vulchanova L, Wang J, Li X, Elde R (1999). Immunocytochemical localization of the vanilloid

receptor 1 (VR1): relationship to neuropeptides, the P2X3 purinoceptor and IB4 binding sites. The


Hegyi Z, Hollo K, Kis G, Mackie K, Antal M (2012). Differential distribution of diacylglycerol lipase-

alpha and N-acylphosphatidylethanolamine-specific phospholipase d immunoreactivity in the

superficial spinal dorsal horn of rats. Glia 60(9): 1316-1329.

Kanai Y, Nakazato E, Fujiuchi A, Hara T, Imai A (2005). Involvement of an increased spinal TRPV1

sensitization through its up-regulation in mechanical allodynia of CCI rats. Neuropharmacology 49(7):

977-984.

Katona I, Freund TF (2008). Endocannabinoid signaling as a synaptic circuit breaker in neurological

disease. Nature medicine 14(9): 923-930.

Kilkenny C, Browne W, Cuthill IC, Emerson M, Altman DG (2010). Animal research: reporting in vivo

experiments: the ARRIVE guidelines. British journal of pharmacology 160(7): 1577-1579.

Kim YH, Back SK, Davies AJ, Jeong H, Jo HJ, Chung G, et al. (2012). TRPV1 in GABAergic interneurons

mediates neuropathic mechanical allodynia and disinhibition of the nociceptive circuitry in the spinal

cord. Neuron 74(4): 640-647.

Kim HI, Kim TH, Shin YK, Lee CS, Park M, Song JH (2005). Anandamide suppression of Na+ currents in

rat dorsal root ganglion neurons. Brain research 1062(1-2): 39-47.

Kwon SG, Roh DH, Yoon SY, Moon JY, Choi SR, Choi HS, et al. (2014). Blockade of peripheral P2Y1

receptors prevents the induction of thermal hyperalgesia via modulation of TRPV1 expression in

carrageenan-induced inflammatory pain rats: involvement of p38 MAPK phosphorylation in DRGs.

Neuropharmacology 79: 368-379.

Lappin SC, Randall AD, Gunthorpe MJ, Morisset V (2006). TRPV1 antagonist, SB-366791, inhibits

glutamatergic synaptic transmission in rat spinal dorsal horn following peripheral inflammation.

European journal of pharmacology 540(1-3): 73-81.

Lever IJ, Robinson M, Cibelli M, Paule C, Santha P, Yee L, et al. (2009). Localization of the

endocannabinoid-degrading enzyme fatty acid amide hydrolase in rat dorsal root ganglion cells and

its regulation after peripheral nerve injury. The Journal of neuroscience : the official journal of the

Society for Neuroscience 29(12): 3766-3780.

Luo H, Cheng J, Han JS, Wan Y (2004). Change of vanilloid receptor 1 expression in dorsal root

ganglion and spinal dorsal horn during inflammatory nociception induced by complete Freund's

adjuvant in rats. Neuroreport 15(4): 655-658.

Malek N, Kucharczyk M, Starowicz K (2014). Alterations in the anandamide metabolism in the

development of neuropathic pain. BioMed research international 2014: 686908.

Page 62 of 76



For Peer Review

23

Mallet C, Dubray C, Duale C (2016). FAAH inhibitors in the limelight, but regrettably. International

journal of clinical pharmacology and therapeutics 54(7): 498-501.

McGrath JC, Lilley E (2015). Implementing guidelines on reporting research using animals (ARRIVE

etc.): new requirements for publication in BJP. British journal of pharmacology 172(13): 3189-3193.

Morisset V, Ahluwalia J, Nagy I, Urban L (2001). Possible mechanisms of cannabinoid-induced

antinociception in the spinal cord. European journal of pharmacology 429(1-3): 93-100.

Morisset V, Urban L (2001). Cannabinoid-induced presynaptic inhibition of glutamatergic EPSCs in

substantia gelatinosa neurons of the rat spinal cord. Journal of neurophysiology 86(1): 40-48.

Nagy I, Friston D, Valente JS, Torres Perez JV, Andreou AP (2014). Pharmacology of the capsaicin

receptor, transient receptor potential vanilloid type-1 ion channel. Progress in drug research.

Fortschritte der Arzneimittelforschung. Progres des recherches pharmaceutiques 68: 39-76.

Nyilas R, Gregg LC, Mackie K, Watanabe M, Zimmer A, Hohmann AG, et al. (2009). Molecular

architecture of endocannabinoid signaling at nociceptive synapses mediating analgesia. The


Okura D, Horishita T, Ueno S, Yanagihara N, Sudo Y, Uezono Y, et al. (2014). The endocannabinoid

anandamide inhibits voltage-gated sodium channels Nav1.2, Nav1.6, Nav1.7, and Nav1.8 in Xenopus

oocytes. Anesthesia and analgesia 118(3): 554-562.

O'Sullivan SE (2007). Cannabinoids go nuclear: evidence for activation of peroxisome proliferator-

activated receptors. British journal of pharmacology 152(5): 576-582.

Park CK, Lu N, Xu ZZ, Liu T, Serhan CN, Ji RR (2011). Resolving TRPV1- and TNF-alpha-mediated spinal

cord synaptic plasticity and inflammatory pain with neuroprotectin D1. The Journal of neuroscience :

the official journal of the Society for Neuroscience 31(42): 15072-15085.

Pernia-Andrade AJ, Kato A, Witschi R, Nyilas R, Katona I, Freund TF, et al. (2009). Spinal

endocannabinoids and CB1 receptors mediate C-fiber-induced heterosynaptic pain sensitization.

Science 325(5941): 760-764.

Ren K, Dubner R (1999). Inflammatory Models of Pain and Hyperalgesia. ILAR journal 40(3): 111-118.

Richardson JD, Kilo S, Hargreaves KM (1998). Cannabinoids reduce hyperalgesia and inflammation via

interaction with peripheral CB1 receptors. Pain 75(1): 111-119.

Snider NT, Walker VJ, Hollenberg PF (2010). Oxidation of the endogenous cannabinoid arachidonoyl

ethanolamide by the cytochrome P450 monooxygenases: physiological and pharmacological

implications. Pharmacological reviews 62(1): 136-154.

Sousa-Valente J, Andreou AP, Urban L, Nagy I (2014a). Transient receptor potential ion channels in

primary sensory neurons as targets for novel analgesics. British journal of pharmacology 171(10):

2508-2527.

Sousa-Valente J, Varga A, Ananthan K, Khajuria A, Nagy I (2014b). Anandamide in primary sensory

neurons: too much of a good thing? The European journal of neuroscience 39(3): 409-418.

Sousa-Valente J, Varga A, Torres Perez JV, Jenes A, Wahba J, Mackie K, et al. (2017). Inflammation of

peripheral tissues and injury to peripheral nerves induce diferring effects in the expression of the

calcium-sensitive anandamide-synthesising enzyme and related molecules in rat primary sensory

neurons. The Journal of comparative neurology.

Page 63 of 76



For Peer Review

24

Spicarova D, Adamek P, Kalynovska N, Mrozkova P, Palecek J (2014a). TRPV1 receptor inhibition

decreases CCL2-induced hyperalgesia. Neuropharmacology 81: 75-84.

Spicarova D, Nerandzic V, Palecek J (2011). Modulation of spinal cord synaptic activity by tumor

necrosis factor alpha in a model of peripheral neuropathy. Journal of neuroinflammation 8: 177.

Spicarova D, Nerandzic V, Palecek J (2014b). Update on the role of spinal cord TRPV1 receptors in

pain modulation. Physiological research / Academia Scientiarum Bohemoslovaca 63 Suppl 1: S225-

236.

Spicarova D, Palecek J (2010). Modulation of AMPA excitatory postsynaptic currents in the spinal

cord dorsal horn neurons by insulin. Neuroscience 166(1): 305-311.

Spicarova D, Palecek J (2009). The role of the TRPV1 endogenous agonist N-Oleoyldopamine in

modulation of nociceptive signaling at the spinal cord level. Journal of neurophysiology 102(1): 234-

243.

Tognetto M, Amadesi S, Harrison S, Creminon C, Trevisani M, Carreras M, et al. (2001). Anandamide

excites central terminals of dorsal root ganglion neurons via vanilloid receptor-1 activation. The

Journal of neuroscience : the official journal of the Society for Neuroscience 21(4): 1104-1109.

Ueda N, Tsuboi K, Uyama T (2013). Metabolism of endocannabinoids and related N-

acylethanolamines: canonical and alternative pathways. The FEBS journal 280(9): 1874-1894.

van der Stelt M, Trevisani M, Vellani V, De Petrocellis L, Schiano Moriello A, Campi B, et al. (2005).

Anandamide acts as an intracellular messenger amplifying Ca2+ influx via TRPV1 channels. The EMBO

journal 24(17): 3026-3037.

Varga A, Jenes A, Marczylo TH, Sousa-Valente J, Chen J, Austin J, et al. (2014). Anandamide produced

by Ca(2+)-insensitive enzymes induces excitation in primary sensory neurons. Pflugers Archiv :

European journal of physiology 466(7): 1421-1435.

Vellani V, Petrosino S, De Petrocellis L, Valenti M, Prandini M, Magherini PC, et al. (2008). Functional

lipidomics. Calcium-independent activation of endocannabinoid/endovanilloid lipid signalling in

sensory neurons by protein kinases C and A and thrombin. Neuropharmacology 55(8): 1274-1279.

Veress G, Meszar Z, Muszil D, Avelino A, Matesz K, Mackie K, et al. (2013). Characterisation of

cannabinoid 1 receptor expression in the perikarya, and peripheral and spinal processes of primary

sensory neurons. Brain structure & function 218(3): 733-750.

Wang J, Okamoto Y, Morishita J, Tsuboi K, Miyatake A, Ueda N (2006). Functional analysis of the

purified anandamide-generating phospholipase D as a member of the metallo-beta-lactamase family.


Wang J, Ueda N (2009). Biology of endocannabinoid synthesis system. Prostaglandins & other lipid

mediators 89(3-4): 112-119.

Wu ZZ, Chen SR, Pan HL (2005). Transient receptor potential vanilloid type 1 activation down-

regulates voltage-gated calcium channels through calcium-dependent calcineurin in sensory neurons.


Yang K, Kumamoto E, Furue H, Li YQ, Yoshimura M (1999). Action of capsaicin on dorsal root-evoked

synaptic transmission to substantia gelatinosa neurons in adult rat spinal cord slices. Brain research

830(2): 268-273.

Page 64 of 76



For Peer Review

25

Zygmunt PM, Petersson J, Andersson DA, Chuang H, Sorgard M, Di Marzo V, et al. (1999). Vanilloid

receptors on sensory nerves mediate the vasodilator action of anandamide. Nature 400(6743): 452-

457.

Page 65 of 76



For Peer Review

26

Figure 1. Anandamide concentration after 20:4-NAPE application in spinal cord slices.

Three different concentrations of 20:4-NAPE (20 µM, 100 µM, 200 µM) were applied on

spinal cord slices. Increasing content of AEA was detected in the extracellular solution after

20:4-NAPE application in a concentration dependent manner (n=5, *p<0.05 compared to

control, #p<0.05 compared to 20 µM and 100 µM 20:4-NAPE application, RM ANOVA on

ranks followed by Student-Newman-Keuls test).

Figure 2. Inhibitory effect of 20:4-NAPE application on excitatory postsynaptic currents

in spinal cord slices from naive animals. (A) An example of native recording of

spontaneous EPSCs from one superficial dorsal horn neuron before (CTRL) and during 20:4-

NAPE (20 µM) application. (B) Application of 20:4-NAPE (20 µM) robustly decreased the

average frequency of sEPSCs (n=13, *p<0.05, Wilcoxon signed-rank test) (C) This is also

evident using cumulative histogram analysis (p<0.05, Kolmogorov-Smirnov test). (D)

Decrease of sEPSC amplitude was not significant using cumulative amplitude analysis. (E)

Recording of dorsal root stimulation-evoked EPSC from one neuron before and during 20:4-

NAPE (20 µM) application. (F) Acute application of 20:4-NAPE (20 µM) significantly

decreased the mean amplitude of eEPSCs (n=15,*p<0.05, Wilcoxon signed-rank test).

Figure 3. The effect of CB1 and TRPV1 receptor antagonists on the 20:4-NAPE-induced

inhibition of sEPSC frequency in naive slices. (A, C) The application of PF514273 (0.2

µM) alone did not change the frequency of sEPSCs significantly (n=11). Following co-

application of PF514273 (0.2 µM) with 20:4-NAPE (20 µM) prevented the 20:4-NAPE

induced inhibition and the sEPSCs frequency did not differ from the control. (B, D) The

application of SB366791 (10 µM, n=10) did not change the sEPSCs frequency. However,

following co-application of SB366791 (10 µM) and 20:4-NAPE (20 µM) a significant

decrease in the sEPSCs frequency was present (*p<0.05 versus control, #p<0.05 versus

SB366791 pretreatment, RM ANOVA on ranks followed by Student-Newman-Keuls test).

(E) The application of both antagonists PF514273 (0.2 µM) with SB366791 (10 µM) did not

change the frequency of sEPSCs significantly (n=8). Subsequent co-application of PF514273

(0.2 µM), SB366791 (10 µM) with 20:4-NAPE (20 µM) prevented the 20:4-NAPE induced

inhibition. (F) The same data are expressed as a percentage of previous recording conditions:

20:4-NAPE (n=13, *p<0.05, Wilcoxon signed-rank test) versus basal frequency of sEPSCs;

PF514273 + 20:4-NAPE (n=11) versus the PF514273 pretreatment; SB366791 + 20:4-NAPE

(n=10, *p<0.05, Wilcoxon signed-rank test) versus the SB366791 pretreatment; PF514237 +

Page 66 of 76



For Peer Review

27

SB366791 + 20:4-NAPE compared to both antagonists pretreatment (n=8). Statistically

significant difference between 20:4-NAPE alone and PF514273 + 20:4-NAPE: #p<0.05, One-

way ANOVA followed by Student-Newman-Keuls test.

Figure 4. Effect of CB1 and TRPV1 antagonists on 20:4-NAPE-induced inhibition of

eEPSC amplitude in naive slices. (A, C) The pretreatment with PF514273 (0.2 µM, n=13)

did not changed the amplitude of the recorded eEPSC in spinal cord slices prepared from

naive animals. Following co-application of PF514273 (0.2 µM) and 20:4-NAPE (20 µM) also

did not significantly changed the amplitude of eEPSC. (B, D) The pretreatment with

SB366791 (10 µM) elicited very similar size of the control eEPSC amplitude. Following co-

application of SB366791 (10 µM) and 20:4-NAPE (20 µM) induced decrease of the eEPSC

amplitude (n=10, *p<0.05 versus control, #p<0.05 versus SB366791 pretreatment, RM

ANOVA on ranks followed by Student-Newman-Keuls test). (E) Data are shown as a

percentage of previous condition to eliminate the effect of antagonist activity, respectively:

20:4-NAPE as percentage of eEPSC basal amplitude (n=15, *p<0.05, Wilcoxon signed-rank

test); PF514273 + 20:4-NAPE as a percentage of PF514273 pretreatment (n=13); SB366791 +

20:4-NAPE as a percentage of SB366791 pretreatment (n=10, *p<0.05, Wilcoxon signed-rank

test).

Figure 5. Application of the 20:4-NAPE decreased the frequency of sEPSCs under

inflammatory conditions. (A) Native recording from one superficial dorsal horn neuron

before and during 20:4-NAPE (20 µM) bath application on spinal cord slice dissected 24 h

after the induction of peripheral inflammation. (B) Application of 20:4-NAPE (20 µM)

significantly decreased the frequency of sEPSCs (n=9, *p<0.05, Wilcoxon signed-rank test).

Figure 6. The effect of CB1 and TRPV1 antagonists on 20:4-NAPE-induced inhibition of

sEPSC frequency under inflammatory conditions: (A, C) The application of PF514273

(0.2 µM, n=16) did not change the frequency of sEPSCs. Subsequent co-application of

PF514273 (0.2 µM) and 20:4-NAPE (20 µM) increased the frequency of sEPSCs without

statistical significance compared to control. (B, D) The frequency of sEPSCs significantly

decreased during application of SB366791 (10 µM, n=16, *p<0.05, RM ANOVA on ranks

followed by Student-Newman-Keuls test). Following co-application of SB366791 (10 µM)

and 20:4-NAPE (20 µM) leaded to even stronger decrease of sEPSC frequency (*p<0.05, RM

ANOVA on ranks followed by Student-Newman-Keuls test). (E) The combined application of

Page 67 of 76



For Peer Review

28

PF514273 (0.2 µM) and SB366791 (10 µM) did not change the frequency of sEPSCs,

subsequent co-application of both antagonists with 20:4-NAPE (20 µM) significantly

decreased the frequency of sEPSCs only compared to control (n=15, *p<0.05, RM ANOVA

on ranks followed by Student-Newman-Keuls test). (F) The same data are shown as a

percentage of previous recording conditions: 20:4-NAPE (n=9) versus sEPSC basal

frequency, PF514237 + 20:4-NAPE versus PF514273 (n=16) pretreatment, SB366791 + 20:4-

NAPE versus SB366791 (n=16) pretreatment and PF514237 + SB366791 + 20:4-NAPE

versus both antagonists pretreatment (n=15). Statistically significant differences: *p<0.05

versus pretreatment, Wilcoxon signed-rank test; #p<0.05 versus PF514237 + 20:4-NAPE co-

application, One-way ANOVA followed by Student-Newman-Keuls test.

Figure 7. Application of 20:4-NAPE decreased the amplitude of evoked EPSCs in

superficial dorsal horn neurons under inflammatory conditions. (A) An example of native

recording from one nociceptive neuron before and during 20:4-NAPE (20 µM) bath

application on acute spinal cord slice prepared 24 h after intraplantar injection of carrageenan.

(B) Acute application of 20:4-NAPE (20 µM) significantly decreased the amplitude of

eEPSCs (n=14; *p<0.05, Wilcoxon signed-rank test).

Figure 8. Antagonists of CB1 and TRPV1 receptors blocked the 20:4-NAPE-induced

decrease of eEPSC amplitude under inflammatory conditions. (A, C) The application of

PF514273 (0.2 µM, n=16) did not changed the amplitude of eEPSCs. Following co-

application of PF514273 (0.2 µM) and 20:4-NAPE (20 µM) slightly increase the amplitude of

eEPSCs without statistical significance. (B, D) The pretreatment with SB366791 (10 µM,

n=11) did not change the amplitude of eEPSC. Following co-application of SB366791 (10

µM) and 20:4-NAPE (20 µM) also did not change the eEPSC amplitude. (E) The same data

are shown as a percentage of previous recording conditions: 20:4-NAPE versus eEPSC basal

amplitude (n=14, *p<0.05, Wilcoxon signed-rank test); PF514237 + 20:4-NAPE versus

PF514273 pretreatment (n=16); SB366791 + 20:4-NAPE versus SB366791 pretreatment

(n=11).

Page 68 of 76



For Peer Review

Figure 1. Anandamide concentration after 20:4-NAPE application in spinal cord slices. Three different concentrations of 20:4-NAPE (20 µM, 100 µM, 200 µM) were applied on spinal cord slices. Increasing

content of AEA was detected in the extracellular solution after 20:4-NAPE application in a concentration dependent manner (n=5, *p<0.05 compared to control, #p<0.05 compared to 20 µM and 100 µM 20:4-

NAPE application, RM ANOVA on ranks followed by Student-Newman-Keuls test).

85x65mm (300 x 300 DPI)

Page 69 of 76



For Peer Review

Figure 2. Inhibitory effect of 20:4-NAPE application on excitatory postsynaptic currents in spinal cord slices from naive animals. (A) An example of native recording of spontaneous EPSCs from one superficial dorsal

horn neuron before (CTRL) and during 20:4-NAPE (20 µM) application. (B) Application of 20:4-NAPE (20 µM)

robustly decreased the average frequency of sEPSCs (n=13, *p<0.05, Wilcoxon signed-rank test) (C) This is also evident using cumulative histogram analysis (p<0.05, Kolmogorov-Smirnov test). (D) Decrease of sEPSC amplitude was not significant using cumulative amplitude analysis. (E) Recording of dorsal root stimulation-evoked EPSC from one neuron before and during 20:4-NAPE (20 µM) application. (F) Acute

application of 20:4-NAPE (20 µM) significantly decreased the mean amplitude of eEPSCs (n=15,*p<0.05, Wilcoxon signed-rank test).

183x212mm (300 x 300 DPI)

Page 70 of 76



For Peer Review

Figure 3. The effect of CB1 and TRPV1 receptor antagonists on the 20:4-NAPE-induced inhibition of sEPSC frequency in naive slices. (A, C) The application of PF514273 (0.2 µM) alone did not change the frequency of

sEPSCs significantly (n=11). Following co-application of PF514273 (0.2 µM) with 20:4-NAPE (20 µM)

prevented the 20:4-NAPE induced inhibition and the sEPSCs frequency did not differ from the control. (B, D) The application of SB366791 (10 µM, n=10) did not change the sEPSCs frequency. However, following co-application of SB366791 (10 µM) and 20:4-NAPE (20 µM) a significant decrease in the sEPSCs frequency was present (*p<0.05 versus control, #p<0.05 versus SB366791 pretreatment, RM ANOVA on ranks

followed by Student-Newman-Keuls test). (E) The application of both antagonists PF514273 (0.2 µM) with SB366791 (10 µM) did not change the frequency of sEPSCs significantly (n=8). Subsequent co-application of

PF514273 (0.2 µM), SB366791 (10 µM) with 20:4-NAPE (20 µM) prevented the 20:4-NAPE induced inhibition. (F) The same data are expressed as a percentage of previous recording conditions: 20:4-NAPE (n=13, *p<0.05, Wilcoxon signed-rank test) versus basal frequency of sEPSCs; PF514273 + 20:4-NAPE

(n=11) versus the PF514273 pretreatment; SB366791 + 20:4-NAPE (n=10, *p<0.05, Wilcoxon signed-rank test) versus the SB366791 pretreatment; PF514237 + SB366791 + 20:4-NAPE compared to both

antagonists pretreatment (n=8). Statistically significant difference between 20:4-NAPE alone and PF514273

Page 71 of 76



For Peer Review

+ 20:4-NAPE: #p<0.05, One-way ANOVA followed by Student-Newman-Keuls test.

177x204mm (300 x 300 DPI)

Page 72 of 76



For Peer Review

Figure 4. Effect of CB1 and TRPV1 antagonists on 20:4-NAPE-induced inhibition of eEPSC amplitude in naive slices. (A, C) The pretreatment with PF514273 (0.2 µM, n=13) did not changed the amplitude of the

recorded eEPSC in spinal cord slices prepared from naive animals. Following co-application of PF514273 (0.2 µM) and 20:4-NAPE (20 µM) also did not significantly changed the amplitude of eEPSC. (B, D) The

pretreatment with SB366791 (10 µM) elicited very similar size of the control eEPSC amplitude. Following co-application of SB366791 (10 µM) and 20:4-NAPE (20 µM) induced decrease of the eEPSC amplitude (n=10, *p<0.05 versus control, #p<0.05 versus SB366791 pretreatment, RM ANOVA on ranks followed by Student-Newman-Keuls test). (E) Data are shown as a percentage of previous condition to eliminate the effect of

antagonist activity, respectively: 20:4-NAPE as percentage of eEPSC basal amplitude (n=15, *p<0.05, Wilcoxon signed-rank test); PF514273 + 20:4-NAPE as a percentage of PF514273 pretreatment (n=13);

SB366791 + 20:4-NAPE as a percentage of SB366791 pretreatment (n=10, *p<0.05, Wilcoxon signed-rank test).

185x96mm (300 x 300 DPI)

Page 73 of 76



For Peer Review

Figure 5. Application of the 20:4-NAPE decreased the frequency of sEPSCs under inflammatory conditions. (A) Native recording from one superficial dorsal horn neuron before and during 20:4-NAPE (20 µM) bath

application on spinal cord slice dissected 24 h after the induction of peripheral inflammation. (B) Application of 20:4-NAPE (20 µM) significantly decreased the frequency of sEPSCs (n=9, *p<0.05, Wilcoxon signed-rank

test).

89x103mm (300 x 300 DPI)

Page 74 of 76



For Peer Review

Figure 6. The effect of CB1 and TRPV1 antagonists on 20:4-NAPE-induced inhibition of sEPSC frequency under inflammatory conditions: (A, C) The application of PF514273 (0.2 µM, n=16) did not change the

frequency of sEPSCs. Subsequent co-application of PF514273 (0.2 µM) and 20:4-NAPE (20 µM) increased the frequency of sEPSCs without statistical significance compared to control. (B, D) The frequency of sEPSCs

significantly decreased during application of SB366791 (10 µM, n=16, *p<0.05, RM ANOVA on ranks followed by Student-Newman-Keuls test). Following co-application of SB366791 (10 µM) and 20:4-NAPE (20

µM) leaded to even stronger decrease of sEPSC frequency (*p<0.05, RM ANOVA on ranks followed by Student-Newman-Keuls test). (E) The combined application of PF514273 (0.2 µM) and SB366791 (10 µM)

did not change the frequency of sEPSCs, subsequent co-application of both antagonists with 20:4-NAPE (20 µM) significantly decreased the frequency of sEPSCs only compared to control (n=15, *p<0.05, RM ANOVA

on ranks followed by Student-Newman-Keuls test). (F) The same data are shown as a percentage of previous recording conditions: 20:4-NAPE (n=9) versus sEPSC basal frequency, PF514237 + 20:4-NAPE

versus PF514273 (n=16) pretreatment, SB366791 + 20:4-NAPE versus SB366791 (n=16) pretreatment and PF514237 + SB366791 + 20:4-NAPE versus both antagonists pretreatment (n=15). Statistically significant differences: *p<0.05 versus pretreatment, Wilcoxon signed-rank test; #p<0.05 versus PF514237 + 20:4-

Page 75 of 76



For Peer Review

NAPE co-application, One-way ANOVA followed by Student-Newman-Keuls test.

178x205mm (300 x 300 DPI)

Page 76 of 76



For Peer Review

Figure 7. Application of 20:4-NAPE decreased the amplitude of evoked EPSCs in superficial dorsal horn neurons under inflammatory conditions. (A) An example of native recording from one nociceptive neuron

before and during 20:4-NAPE (20 µM) bath application on acute spinal cord slice prepared 24 h after

intraplantar injection of carrageenan. (B) Acute application of 20:4-NAPE (20 µM) significantly decreased the amplitude of eEPSCs (n=14; *p<0.05, Wilcoxon signed-rank test).

89x117mm (300 x 300 DPI)

Page 77 of 76



For Peer Review

Figure 8. Antagonists of CB1 and TRPV1 receptors blocked the 20:4-NAPE-induced decrease of eEPSC amplitude under inflammatory conditions. (A, C) The application of PF514273 (0.2 µM, n=16) did not

changed the amplitude of eEPSCs. Following co-application of PF514273 (0.2 µM) and 20:4-NAPE (20 µM) slightly increase the amplitude of eEPSCs without statistical significance. (B, D) The pretreatment with

SB366791 (10 µM, n=11) did not change the amplitude of eEPSC. Following co-application of SB366791 (10 µM) and 20:4-NAPE (20 µM) also did not change the eEPSC amplitude. (E) The same data are shown as a percentage of previous recording conditions: 20:4-NAPE versus eEPSC basal amplitude (n=14, *p<0.05, Wilcoxon signed-rank test); PF514237 + 20:4-NAPE versus PF514273 pretreatment (n=16); SB366791 +

20:4-NAPE versus SB366791 pretreatment (n=11).

184x99mm (300 x 300 DPI)

Page 78 of 76



Documents

For Peer Review - Imperial College London...The paper “Peripheral inflammation alters endocannabinoid-induced modulation of nociceptive spinal dorsal horn synaptic transmission”