|Research Focus
Sensation is painless
Miriam B. Goodman
Department of Molecular and Cellular Physiology, Stanford University, Stanford, CA 94305, USA
Emily Dickinson declared: `After great pain, a formal
feeling comes'. This formal feeling begins when sensory
neurons are activated by noxious stimuli, such as step-
ping on a tack. Recently, Seymour Benzer's group ident-
ified sensory neurons in Drosophila larvae that mediate
aversive responses to noxious heat and mechanical
stimuli. Thresholds for behavioral and nerve responses
are elevated by mutations in the painless gene, which
encodes a TRP ion channel protein. Painless thus joins
an elite group of TRPs implicated in sensory transduc-
tion in insects, nematodes, mammals and fish.
Sensation continuously guides and modifies behavior,
assuring that animals are able to find food and mates,
to move toward favorable environments, and to avoid
predators. Information about the external world is carried
by light, odors, sounds, touches, temperature and even
electric fields. Sensory cells specialized to detect each kind
of stimulus respond to these cues and provide animals
with qualitative, quantitative and spatial information
about the world. Many animals also have sensory neurons
activated preferentially by stimuli with the potential to
cause tissue damage. In humans and other mammals,
such sensory neurons are called nociceptors and generate
signals interpreted by the CNS as pain.
A crucial step in sensation is the activation of
`transducer' channels, which open in proportion to the
strength of the stimulus. Such channels can be activated
indirectly, as in visual transduction, or directly, as hypo-
thesized for hearing [1]. Only now are we beginning to
uncover potential transducer channels needed for hearing,
touch, temperature and pain sensation. Progress in this
arena has accelerated through the use of genetic screens
in the nematode Caenorhabditis elegans, Drosophila fruit
flies, and the zebrafish Danio rerio [2]. Two classes of ion
channel proteins have emerged from these studies as
potential mechanotransducer channels: DEG/ENaCs and
TRPs. (The DEG/ENaC superfamily is named after the
first two sub-families identified: degenerins, C. elegans
genes that when mutated can cause cellular degeneration,
and ENaCs, subunits of vertebrate epithelial Na
chan-
nels. TRP channels are named after the product of the
transient receptor potential gene in Drosophila). This
article highlights Painless, a newly characterized member
of the TRPN subfamily of TRP channels required for
sensation of noxious thermal and mechanical stimuli [3],
and will discuss its sisters, no-mechanoreceptor-potential
C (NompC) [4,5] and the ankyrin-like ANKTM1 [6,7].
Naked neurons and nociceptors
The epidermis of Drosophila larvae is densely innervated
by sensory dendrites from type II or multidendritic (md)
sensory neurons [8]. These cells, which are not clearly
associated with accessory structures, display morpho-
logical similarities to vertebrate sensory neurons that also
terminate in free nerve endings embedded in the skin. The
precise function of md sensory neurons was unknown until
recently. The elegant work of Tracey et al. [3] showed
that signaling by md-da neurons is required for normal
responses to noxious thermal (T . 41 8C) and mechanical
(F . 45 mN) stimuli in Drosophila larvae (Figure 1a).
Thus md-da neurons, or a subset of them, are likely to be
nociceptors. Consistent with this notion, most if not all
md-da neurons are thermosensitive [9]. It should be noted,
however, that responses to both thermal and mechanical
stimuli have yet to be measured in individual md-da
neurons, leaving open the question of whether thermal
and mechanical stimuli are detected by distinct md-da
neurons or whether individual md-da neurons are poly-
modal sensory neurons that can detect both kinds of
stimuli. Additional mechanosensory neurons are present
Figure 1. Responses to noxious stimuli in Drosophila larvae depend on intact
multidendritic md-da neurons and wild-type painless. (a) The response to noxious
thermal (T . 41 8C) and mechanical (F . 45 mN) stimuli. Wild-type animals rapidly
(latency ,1 s) bend and roll away from the site of stimulation. Genetic screens for
animals that fail to respond to noxious heat have uncovered several genes needed
for this robust, aversive behavior. The painless gene, which is predicted to encode
a transient receptor potential (TRP) ion channel, is the first of these genes to be
cloned. This behavior also requires signaling from md-da neurons because
responses to heat are eliminated by targeted expression of tetanus toxin. (b) Pain-
less protein is expressed in the sensory dendrites of md-da neurons and concen-
trated in puncta. Taken together, these results support the idea that Painless
contributes to a sensory transduction channel; Painless puncta might represent
sites of sensory transduction.
(b)(a)
TRENDS in Neurosciences
Corresponding author: Miriam B. Goodman (mbgoodman@stanford.edu).
TINS 133
Update TRENDS in Neurosciences Vol.not known No.not known Month 0000 1
ARTICLE IN PRESS
http://tins.trends.com 0166-2236/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.tins.2003.09.013
in Drosophila larvae, because the md-da neurons are
not required for responses to non-noxious mechanical
stimuli. It will be interesting to learn how sensory
thresholds vary among md-da neurons and whether such
functional variation correlates with recently described
anatomical subtypes [10].
Polymodal Painless?
A nociceptor is a sensory neuron that responds to diverse
stimuli that have the potential to cause tissue damage.
This operational definition, originally proposed by
Sherrington [11], implies that nociceptors are polymodal
­ that is, they have the ability to detect noxious thermal,
mechanical and/or chemical stimuli. To accomplish this
feat, polymodal sensory neurons could express a team
of specialists, detectors that respond to only one kind
of stimulus. Alternatively, one or a few generalists ­
detectors activated by multiple kinds of stimuli ­ could
operate in polymodal sensory neurons. Painless could be
one such generalist detector, a potential transducer of both
thermal and mechanical stimuli. In the absence of the
painless gene, larvae fail to roll away from thermal and
mechanical stimuli that reliably elicit such responses in
wild-type animals. In painless mutants, heating to .41 8C
fails to increase the firing rate in one class of temperature-
sensitive units in the abdominal nerve (which contains the
axons of md-da neurons). Moreover, Painless protein is
localized to md-da dendrites, where it is concentrated in
puncta (Figure 1b). An exciting possibility is that Painless
puncta represent specialized sites of sensory transduction
in md-da dendrites.
Other TRPNs, more sensations...
The generalist Painless is closely related at the molecular
level to two additional TRP channels that appear to be
more specialized. These two channels, ANKTM1 in mice
and flies and NompC in flies and zebrafish, were identified
using molecular and genetic approaches and are proposed
to transduce thermal and mechanical stimuli, respectively.
The first of these channels, mouse ANKTM1, is acti-
vated by noxious cold (threshold T , 17 8C) and by the
cooling agent, icilin [6]. It is coexpressed with another
TRP channel, TRPV1, in a subset of sensory neurons in
mammalian dorsal root ganglia. TRPV1 is a generalist
detector, activated by noxious heat, pH, vanilloids and
endocannabinoids [12]. The subset of dorsal root ganglion
neurons that expresses both ANKTM1 and TRPV1 is
proposed to comprise polymodal nociceptors [6], although
this awaits confirmation in functional studies. Paradoxi-
cally, Drosophila ANKTM1 (34% identical to mouse
ANKTM1) is activated by heating (threshold T . 24 8C)
and not by cooling [7]. The function of Drosophila
ANKTM1 and its expression pattern are unknown. None-
theless, it is easy to imagine exploring the molecular basis
for temperature detection by constructing chimeras of
these two proteins.
A second TRPN channel related to Painless, NompC, is
needed to respond to mechanical stimuli applied to sensory
bristles covering the body of adult Drosophila fruit flies [4].
Bristles are innervated by ciliated or type I sensory
neurons, which are believed to express NompC. Such
sensory neurons display striking similarities to vertebrate
hair cells, specialized cells that detect sound and vibration
in vertebrate inner ears. For instance, the apical surfaces
of both types of sensory cells are bathed in an unusual
extracellular solution that is high in K
[13,14]. Although
much is known about transducer currents in vertebrate
hair cells [15,16], the molecular identity of transducer
channels has remained obscure. A recent breakthrough is
the identification of an ortholog of NompC in zebrafish [5].
The zebrafish NompC gene is expressed by hair cells in the
inner ear and lateral line sensory organs that sense water
flow across the body surface. Consistent with a role for
NompC in sensory mechanotransduction by hair cells,
inactivating NompC eliminated acoustic startle responses
in zebrafish larvae and reduced sensory microphonic
potentials recorded in lateral line organs. With this report,
the similarity between fly bristle mechanoreceptors and
vertebrate hair cells extends to the molecular level.
Future directions
It is assumed, but not proven, that ion channels are direct
transducers of thermal and mechanical stimuli. A first test
of this idea is to determine the effect of loss-of-function
mutations on sensory currents measured in vivo. This
test has yet to be conducted for Painless or ANKTM1.
Extracellular measurements in Drosophila bristles [4] and
zebrafish lateral-line organs [5] are consistent with the
idea that NompC is crucial for sensory mechanotrans-
duction. A residual mechanoreceptor current is present in
Drosophila bristles lacking NompC, however, suggesting
that it is not the only transducer operating in sensory
bristles. A second test is to show that each channel
responds directly to thermal and/or mechanical stimuli
in heterologous cells. This test has been conducted for
thermal sensitivity of mouse and fly ANKTM1 [6,7].
Analogous tests for mechanical sensitivity are proble-
matic, however, because specialized cellular structures
might be essential for gating. Osmotic swelling has been
used as a surrogate mechanical stimulus for TRPV
channels implicated in sensory mechanotransduction in
mammals [17] and flies [18]. These data show that these
channels can respond to mechanical stimuli but it is
unclear how activation by osmotic stimulation is related to
how channels are activated in vivo.
Exactly how transducer channels respond to thermal
and/or mechanical stimuli remains an enticing mystery.
All TRPN channels contain a tandem array of ankyrin
repeats in their N terminals. These domains could anchor
TRPN channels to specialized elements in the cytoskele-
ton, as proposed for NompC [4]. Anchors in sensory cells
could regulate whether or not a given TRPN channel is
activated by mechanical stimuli. Such channels would
have the potential to transduce mechanical energy but
would do so only when expressed in a cell that provides a
suitable anchor. In this scenario, a generalist such as
Painless might be sensitive to both thermal and mechan-
ical stimuli in some neurons and detect only thermal
stimuli in others. The importance of cellular context could
be tested, for example, by expressing Painless in NompC-
expressing cells (and vice versa). Sensitivity to diverse
stimuli could also arise from combinatorial expression of
TINS 133
Update TRENDS in Neurosciences Vol.not known No.not known Month 00002
ARTICLE IN PRESS
http://tins.trends.com
multiple TRPs, as suggested for TRPVs in C. elegans [19].
The discovery that md-da neurons and Painless are essen-
tial for nociception in Drosophila larvae introduces the
idea that sensory thermotransduction and mechanotrans-
duction could operate via a single transducer channel.
Acknowledgements
I thank J. Art, T. Clandinin and D. Lenzi for comments. Research in my
laboratory is supported by the Alfred P. Sloan Foundation, the Donald E.
and Delia B. Baxter Foundation, and a Terman award from Stanford
University.
References
1 Corey, D.P. and Hudspeth, A.J. (1979) Response latency of vertebrate
hair cells. Biophys. J. 26, 499­506
2 Ernstrom, G.G. and Chalfie, M. (2002) Genetics of sensory mechano-
transduction. Annu. Rev. Genet. 36, 411­453
3 Tracey, W.D. et al. (2003) painless, a Drosophila gene essential for
nociception. Cell 113, 261­273
4 Walker, R.G. et al. (2000) A Drosophila mechanosensory transduction
channel. Science 287, 2229­2234
5 Sidi, S. et al. (2003) NompC TRP channel required for vertebrate
sensory hair cell mechanotransduction. Science 301, 96­99
6 Story, G.M. et al. (2003) ANKTM1, a TRP-like channel expressed
in nociceptive neurons, is activated by cold temperatures. Cell 112,
819­829
7 Viswanath, V. et al. (2003) Opposite thermosensor in fruitfly and
mouse. Nature 423, 822­823
8 Bodmer, R. and Jan, Y.N. (1987) Morphological differentiation of the
embryonic peripheral neurons in Drosophila. Roux Arch. Dev. Biol.
196, 69­77
9 Liu, L. et al. (2003) Identification and function of thermosensory
neurons in Drosophila larvae. Nat. Neurosci. 6, 267­273
10 Grueber, W.B. et al. (2002) Tiling of the Drosophila epidermis by
multidendritic sensory neurons. Development 129, 2867­2878
11 Sherrington, C.S. (1904) Qualitative difference of spinal reflex
corresponding with qualitative difference of cutaneous stimulus.
J. Physiol. 30, 39­46
12 Patapoutian, A. et al. (2003) ThermoTRP channels and beyond:
mechanisms of temperature sensation. Nat. Rev. Neurosci. 4,
529­539
13 Grunert, U. and Gnatzy, W. (1987) K
and Ca2
in the receptor lymph
of arthropod cuticular mechanoreceptors. J. Comp. Physiol. A 161,
329­333
14 Johnstone, C.G. et al. (1963) Sodium and potassium in vertebrate
cochlear endolymph as determined by flame microspectrophotometry.
Comp. Biochem. Physiol. 9, 335­341
15 Fettiplace, R. and Fuchs, P.A. (1999) Mechanisms of hair cell tuning.
Annu. Rev. Physiol. 61, 809­834
16 Pickles, J.O. and Corey, D.P. (1992) Mechanoelectrical transduction by
hair cells. Trends Neurosci. 15, 254­259
17 Liedtke, W. et al. (2000) Vanilloid receptor-related osmotically
activated channel (VR-OAC), a candidate vertebrate osmoreceptor.
Cell 103, 525­535
18 Kim, J. et al. (2003) A TRPV family ion channel required for hearing
in Drosophila. Nature 424, 81­84
19 Tobin, D. et al. (2002) Combinatorial expression of TRPV channel
proteins defines their sensory functions and subcellular localization in
C. elegans neurons. Neuron 35, 307­318
TINS 133
Update TRENDS in Neurosciences Vol.not known No.not known Month 0000 3
ARTICLE IN PRESS
http://tins.trends.com