In mammals, the primary afferent neurons located in dorsal root ganglion (DRG) deliver the peripheral sensory information via the dorsal root (or posterior root) to the spinal cord with three types of afferent fibers, the Aß-fibers, Aδ-fibers, and C-fibers [3] [4] [5].

Aß-fibers relay the tactile information, large in diameter and low in activation thresholds [3] [4] [5]. Both Aδ- and C-fibers possess nociceptors [4] [5], while relay the mechanical, thermal, nociceptive and chemical stimuli, small in diameter and high in activation thresholds [3] [4] [5]. The Aδ-fibers form synapses onto the laminae I, II and V of spinal dorsal horn [5] [6], while the C-fibers onto the laminae I and II [5] [6]. The laminae I, II and V in spinal dorsal horn are responsible for relaying the nociceptive inputs.

Many nociceptive neurons in lamina I of spinal dorsal horn fruther project to several brain regions, such as the caudal ventrolateral medulla (CVLM), periaqueductal grey (PAG), parabrachial area (PB) and so on [5] [6]. The PAG neurons can activate the neurons of rostral medullary raphe nuclei [7] [8], while the raphe serotonergic (5-HT) neurons in turn project to the spinal nociceptive neurons and inhibit their activities, suppressing onto the spinal pain transmission and completing the feedback circuits for pain transmission in spinal cord [5] [7] [8].

On May 6, 2016, the US athletes of NBA in television proposed the nociceptive-sympathetic coupling hypothesis for pain sensations [1] [2]. The sympathetically maintained pain such as causalgia and reflex sympathetic dystrophy can support the nociceptive-sympathetic coupling [9] [10].

Recently, Cai wrote an article depicting the neural pathways for the nociceptive-sympathetic coupling [2]. One the one hand, it was shown that the muscular sympathetic nerve was coupled in activity to PAG notably and a few others in brain for sustained muscle pain [11], the tonic glutamatergic inputs from PAG to rostral ventrolateral medulla (RVLM) were increased for the high blood pressure of neuropathic pain [12], and the cardiac sympathetic activation eliciting pain activated the PAG, PB, dorsal raphe, and locus coeruleus (LC) in brain [13], all demonstrating the links of brain PAG to the sympathetic effects of pain [2]. On the other hand, more importantly but neglected widely by most pain researchers nowadays, it was demonstrated that not only the sympathetic preganglionic neurons received afferents directly from the nociceptive lamina I [14] and V [15] in spinal dorsal horn, but also could be activated by these spinal nociceptive neurons [16] [17], direct completing the nociceptive-sympathetic coupling at the spinal cord [2].

In addition to the nociceptive-sympathetic coupling, it was also depicted the nociceptive-respiratory coupling [2] [18], which was mediated via PB in brain rather than the spinal cord or brain PAG.

In brief, the direct sympathetic links from the nociceptive neurons in laminae I and V of dorsal spinal horn as well as the nociceptive neurons in periaqueductal grey (PAG) of brain complete the neural pathways for nociceptive-sympa- thetic coupling in vertebrates.

The neural pathways for nociceptive-sympathetic coupling were revealed in mammals [2] including humans. To consider the evolutionary perspectives of nociceptive-sympathetic coupling, it is certainly necessary to first consider the presence of pain sensations in other animals than mammals.

The spinal cord, PAG and PB for nociceptive-sympathetic coupling and nociceptive-respiratory coupling are all located in the central nervous system below thalamus, and are ancient and primitive in vertebrates during evolution [2]. Indeed, pain sensations exist in various vertebrates, including the early fishes [2] [19] [20] [21]. Neuronal responses have been shown in spinal cord, cerebellum, tectum, and telencephalon in both goldfish and trout for mechanoceptive brush and nociceptive pin-prod [21], indicating both spinal cord and brain responsible for pain sensations. Whereas, because the lung has emerged after amphibians while fishes adopt the gill to breathe, the nociceptive-respiratory coupling via PB neurons may be different in fishes [2].

In the step further, it is certainly necessary to consider the presence of pain sensations in animals more ancient and primitive than vertebrates. Because of the variations of central neural structures in invertebrates and primitive vertebrates, the biochemical perception and neural transmission of nociception become the key marks for assaying the pain sensations in these ancient and primitive animals.

Nociceptors or nociceptive neural transmission has been discovered throughout the Animalia kingdom, such as Annelida [22] [23], Mollusca including Aplysia [22] [23], Nematoda including C elegans [23] [24], Arthropoda including Drosophila [22] [23], Cephalochordata including amphioxus [25], Agnatha including lamprey [26], Vertebrata including fishes [2] [19] [20] [21] and mammals [5] [7] [8], and so on. Therefore, pain sensations are widely present in almost all animals, even including those emerging early in evolution at least as ancient and primitive as Annelida.

In invertebrates and primitive vertebrates, the evolutionary variations in both brain complexity and autonomic regulation occur in various primitive animals, making them dramatically different from those in higher vertebrates. In this regard, the neural structures employed in nociceptive-sympathetic coupling may not be found in these primitive animals, just as the nociceptive-respiratory coupling via PB neurons may be different in fishes lack of lung [2]. Because of such discrepancies, to reveal the pain mechanisms in invertebrates and primitive vertebrates, it is necessary to transform the nociceptive-sympathetic coupling mechanism into more general forms.

The nociceptive-sympathetic coupling in vertebrates comprises the direct coupling of the nociceptive neurons in laminae I and V of dorsal spinal horn to the sympathetic output [2], and the coupling of nociceptive neurons in periaqueductal grey (PAG) of brain to the sympathetic effects [2]. The spinal cord and PAG are specific neural structures in vertebrates, belonging to the peripheral somite and anterior brain respectively. On the other hand, the peripheral somite and anterior brain are general structures present in all vertebrates and most invertebrates more advanced than those in Annelida. In this regard, to correspond and compare the nociceptive-sympathetic coupling in vertebrates to the pain mechanism in invertebrates, it is neccesary to make use of the more general structures as peripheral somite and anterior brain rather than spinal cord and PAG respectively. Accordingly, it is herein classified the nociceptive-sympathetic coupling for pain sensations from spinal cord as the somite nociceptive coupling, and the nociceptive-sympathetic coupling from PAG of brain as the brain nociceptive coupling, respectively (Table 1). With the peripheral somite and anterior brain as general structures present in all vertebrates and most invertebrates more advanced than those in Annelida, it is possible to consider the suitability for the somite and brain nociceptive coupling in primitive vertebrates and invertebrates.

Updated progressions from abundant investigations of noxious perceptions on several model animals would help determine the suitability for the somite and brain nociceptive coupling in primitive animals, notably in amphioxus, Drosophila and C elegans.

1) Amphioxus: Amphioxus is important in that it is the key animal evolving from invertebrates to vertebrates. Unfortunately, the nociceptive research in amphioxus is very few. Up to now, only gene analysis of Rfamide neuropeptide suggested its plausible nociceptive function [25]. Whereas, because nociceptions have been identified in such invertebrates as Drosophila [22] [23] and C elegans [23] [24] as in vertebrates, it is certain that the nociception must also exist in amphioxus between the invertebrates and vertebrates.

2) Drosophila: Extensive investigations have been performed in Drosophila for revealing the neural mechanism of nociceptive reactions in this animal species [22] [23]. Notably, it was reported that the class IV dendrite arborization (C4 da) sensory neurons in the peripheral nervous system were responsible for perception of multiple nociceptive modalities, and then activated the segmentally arrayed local interneurons (medial clusters of C4 da second-order interneurons [mCSIs]) in the ventral nerve cord that were necessary and sufficient to trigger rolling behavior [27]. Likewise, it was also shown that a population of interneurons in the nerve cord of Drosophila, termed Down-and-Back (DnB)

neurons, were activated by noxious heat, promoted nociceptive behavior, and were required for robust escape responses to noxious stimuli [28]. Obviously, the interneurons in the nerve cord of Drosophila in both of the two reports corresponded to the somite nociceptive coupling. In a more recent report, complex reactions to graded encodings of noxious stimuli relied on the neural circuits across segments [29]. More investigations are required to demonstrate the inter-segmental brain nociceptive coupling in Drosophila.

3) C elegans: There have also been many investigations in C elegans for nociceptive reactions in this simple primitive animal species [23] [24]. Especially, the noxious sensory neuron PVD and interneuron ALA are both long, covering the majority of body length of the small C elegans, with the interactions between PVD and ALA across several segments [30]. Obviously, the somite nociceptive coupling exists in the PVD and ALA coupling, while whether the anterior portion of ALA lack of PVD inputs can transfer noxious stimuli as brain nociceptive coupling to coordinate behavioral responses deserves more investigations.

In all, delicate researches on the neural circuits for responses to noxious stimuli in both Drosophila and C elegans have confirmed the widespread occurrence of somite nociceptive coupling in invertebrates, while the brain nociceptive coupling in invertebrates requires more investigations.