Pain Pathways and Central Modulation

Nociceptive information enters the central nervous system through the dorsal roots and terminates in the dorsal horn of the spinal cord gray matter, organized into ten cytoarchitectonic layers called Rexed laminae. Aδ fibers, which carry fast, sharp "first pain," terminate predominantly in lamina I (marginal zone) and lamina V. Unmyelinated C fibers, carrying slow, burning "second pain," terminate mainly in lamina II (substantia gelatinosa) and also lamina I. From here, second-order projection neurons — many of them wide-dynamic-range (WDR) neurons that also receive innocuous touch input — send axons that cross the midline in the anterior white commissure, usually within one to two spinal segments of entry, and ascend in the contralateral anterolateral white matter as the spinothalamic tract, the principal component of the broader anterolateral system. This crossed, ascending organization is the single most clinically important anatomical fact in this article: a spinal cord lesion that damages the anterolateral quadrant on one side produces loss of pain and temperature sensation on the opposite side of the body below the lesion, while touch (carried in the uncrossed dorsal columns) is preserved — the classic dissociated sensory loss of Brown-Séquard syndrome. This double dissociation, seen at the bedside, is itself a form of evidence for the anatomical separation of pain/temperature pathways from touch/proprioception pathways. Crucially, pain is not simply "transmitted" through this circuit — it is gated. The dorsal horn contains local inhibitory interneurons that can suppress transmission from nociceptors onto projection neurons, and these interneurons are themselves controlled both by incoming large-fiber touch input (the substrate of gate control theory) and by descending fibers from the brainstem that release serotonin, norepinephrine, and endogenous opioid peptides. The same anatomical node — the dorsal horn synapse between primary afferent and projection neuron — is thus a convergence point for ascending sensory drive, local segmental inhibition, and descending central control. This convergence is why psychological state, attention, prior conditioning, and pharmacology can all change how much a given injury hurts.

Step 1 — Peripheral transduction and conduction. Nociceptors (covered in Nociception and Pain Receptors) transduce noxious mechanical, thermal, or chemical stimuli into receptor potentials and then action potentials. Two fiber types matter here:

The roughly 15- to 50-fold conduction velocity difference between Aδ and C fibers is precisely why a stubbed toe produces two temporally separated sensations: a sharp jab arriving first, followed half a second or so later by a duller ache.

Step 2 — Dorsal horn synapse and the gate. Primary afferents release glutamate (fast transmission) and, especially from C fibers, neuropeptides such as substance P and calcitonin gene-related peptide (CGRP), which act on projection neurons and modulate synaptic gain over slower timescales. Gate control theory, proposed by Ronald Melzack and Patrick Wall in 1965, posited a dorsal horn "gating" circuit in the substantia gelatinosa: small-fiber (nociceptive) input tends to open the gate (facilitating transmission to projection neurons), while large-fiber (Aβ, touch) input tends to close it, by exciting inhibitory interneurons that presynaptically and postsynaptically inhibit the projection neuron. This is why rubbing or shaking an injured area — flooding the dorsal horn with large-fiber touch input — reduces perceived pain, and it is the physiological rationale for transcutaneous electrical nerve stimulation (TENS).

Step 3 — Ascending transmission: two parallel limbs. Second-order projection neurons, mostly in laminae I and V, send axons across the midline and ascend as the spinothalamic tract, which anatomically fractionates into two functionally distinct components:

• Lateral (neo)spinothalamic tract — fast, somatotopically organized, projects to the ventral posterolateral (VPL) nucleus of the thalamus, which relays to primary and secondary somatosensory cortex (S1/S2). This limb supports the sensory-discriminative dimension of pain: location, intensity, quality.
• Medial (paleo)spinothalamic and spinoreticular pathways — slower, more diffusely organized, project to intralaminar thalamic nuclei and also synapse in the reticular formation and periaqueductal gray en route. These relay onward to the anterior cingulate cortex, insula, and prefrontal cortex, supporting the affective-motivational dimension of pain — the unpleasantness, the urge to escape, the emotional coloring.

This anatomical split explains a clinically important dissociation: cingulotomy or cingulate damage can leave patients able to localize a painful stimulus accurately while reporting that it no longer bothers them — pain asymbolia, where the sensory-discriminative channel survives but the affective channel is disrupted.

Step 4 — Descending modulation: closing the gate from above. The brainstem does not merely receive pain signals passively; it actively controls how much of the incoming signal gets through. The key circuit is the periaqueductal gray (PAG) → rostral ventromedial medulla (RVM) → dorsal horn pathway:

• The PAG, in the midbrain surrounding the cerebral aqueduct, receives input from the hypothalamus, amygdala, and prefrontal cortex — structures encoding threat, context, and expectation.
• PAG neurons project to the RVM (including the nucleus raphe magnus), which sends serotonergic (5-HT) and adjacent noradrenergic (from the locus coeruleus) axons down the dorsolateral funiculus to the dorsal horn.
• These descending fibers synapse onto inhibitory interneurons and directly onto projection neurons, suppressing nociceptive transmission at the first central synapse — the same node targeted by the gate control mechanism.
• The RVM contains two functionally opposed cell classes: "off-cells," which pause firing just before a nociceptive reflex and whose activity is analgesic, and "on-cells," which increase firing before a nociceptive reflex and facilitate pain. Opioids act partly by inhibiting on-cells and disinhibiting off-cells.

Step 5 — Endogenous opioids and receptor pharmacology. The brain produces its own opioid peptides — enkephalins, β-endorphin, and dynorphins — derived from three distinct precursor proteins (proenkephalin, proopiomelanocortin, and prodynorphin respectively). These peptides act on three G-protein-coupled receptor types:

Opioid receptors are Gi/o-coupled: activation inhibits adenylyl cyclase, opens G-protein-coupled inwardly rectifying K⁺ (GIRK) channels (hyperpolarizing the neuron), and closes voltage-gated Ca²⁺ channels, reducing neurotransmitter release. At the dorsal horn, presynaptic μ-receptors on primary afferent terminals reduce glutamate and substance P release directly onto projection neurons — a mechanism distinct from, but converging on, the same synapse targeted by gate control and descending serotonergic/noradrenergic modulation. In the PAG, opioids disinhibit output neurons by suppressing local GABAergic interneurons, activating the descending analgesic pathway indirectly. Exogenous opioids (morphine, fentanyl, oxycodone) are exogenous agonists at these same receptors, chiefly μ, which is why they produce analgesia but also share the receptor's other functions — respiratory depression via brainstem μ-receptors, constipation via enteric μ-receptors, and reward via mesolimbic dopamine circuits, the latter underlying addiction liability.

Gate control theory — Melzack and Wall (1965). Before 1965, pain theory was dominated by a simple "specificity" model: dedicated pain fibers carried a fixed-intensity signal straight to a "pain center." This could not explain everyday observations — that rubbing an injury reduces pain, that soldiers with severe wounds sometimes feel little pain in combat, or that the same stimulus hurts more or less depending on attention and anxiety. Ronald Melzack and Patrick Wall published "Pain Mechanisms: A New Theory" in Science in 1965, proposing that a spinal gating mechanism in the substantia gelatinosa modulates the balance of large-fiber and small-fiber input before it reaches central "transmission" (T) cells, and further proposed that descending influences from the brain could adjust this gate. The theory's specific circuit diagram has required revision — later physiology showed the substantia gelatinosa circuitry is more complex than the original wiring diagram, with multiple interneuron types rather than a single inhibitory population — but the central claims have held up robustly: large-fiber input can inhibit dorsal horn nociceptive transmission, and descending pathways can gate incoming pain signals. This theory directly motivated TENS therapy and reframed pain as an actively regulated CNS process rather than a passive readout, arguably the most influential reconceptualization in the history of pain research.

Stimulation-produced analgesia — Reynolds (1969) and Liebeskind's group (1970s). David Reynolds demonstrated that electrical stimulation of the periaqueductal gray in rats produced analgesia sufficient to allow abdominal surgery without additional anesthetic, while the animals remained otherwise alert and responsive. Subsequent work by John Liebeskind and colleagues confirmed and extended this "stimulation-produced analgesia" (SPA), showing it was anatomically specific to the PAG and adjacent midbrain structures, and — critically — that it could be substantially reversed by the opioid antagonist naloxone. This naloxone-reversibility was the key piece of evidence linking electrical stimulation of a discrete brain site to the endogenous opioid system, rather than to some generic distraction or stress effect, and it directly predicted the existence of an endogenous ligand for the opioid receptor before that ligand (enkephalin, isolated by Hughes and Kosterlitz in 1975) was chemically identified.

Opioid receptor binding and the discovery of endogenous opioids — Pert and Snyder (1973); Hughes and Kosterlitz (1975). Candace Pert and Solomon Snyder used radiolabeled opioid ligands to demonstrate stereospecific, saturable opioid binding sites in mammalian brain tissue, concentrated in regions including the PAG, thalamus, and dorsal horn — direct biochemical proof that the brain has dedicated opioid receptors, published in Science in 1973. This raised an obvious question: why would the brain evolve receptors for a poppy alkaloid? The answer came in 1975, when John Hughes and Hans Kosterlitz isolated and sequenced two endogenous pentapeptides from pig brain — methionine-enkephalin and leucine-enkephalin — the first identified endogenous opioids, showing that morphine and its relatives act by mimicking a natural neurochemical system rather than by some pharmacological coincidence. Together, these findings converted gate control theory's abstract "descending inhibition" into a defined molecular and anatomical circuit: PAG → RVM → dorsal horn, gated by endogenous opioid peptides acting on real, identifiable receptors.

Naloxone reversal as the definitive causal test. Across many subsequent studies, the reproducible finding that naloxone — a pure opioid antagonist with no other major pharmacological action at clinical doses — blocks analgesia produced by stress, placebo administration, acupuncture-like stimulation, and PAG/RVM stimulation, provided a clean causal test distinguishing opioid-mediated analgesia from other mechanisms. This pharmacological "on/off switch" logic (block the receptor, lose the effect; the effect must therefore depend on that receptor) remains the standard evidentiary bar in pain neuroscience, analogous to how receptor antagonists are used to establish causal neurotransmitter roles throughout neuropharmacology.

Example 1 — Localizing a spinal cord lesion from a sensory exam. A 34-year-old presents after a knife wound to the right side of the spinal cord at approximately the T10 level. Examination reveals: loss of pain and temperature sensation on the left side of the body below approximately T12; loss of touch, vibration, and proprioception, plus ipsilateral motor weakness, on the right side below the lesion. Reasoning: The dorsal columns (touch, vibration, proprioception) ascend ipsilaterally and do not cross until the medulla, so damage to the right dorsal column produces right-sided (ipsilateral) deficits — consistent with the exam. The corticospinal tract also has not yet crossed at this level (it crossed in the medulla), so right-sided motor weakness is also ipsilateral to the lesion — consistent. The spinothalamic tract, by contrast, carries fibers that crossed the midline one to two segments below their entry point, so fibers ascending on the right side of the cord at T10 originated from dermatomes on the left side of the body. Damage to the right anterolateral column therefore produces contralateral (left-sided) loss of pain and temperature a few segments below the lesion. This pattern — ipsilateral touch/motor loss plus contralateral pain/temperature loss — is the classic Brown-Séquard syndrome, and the two- to three-segment offset between the motor/touch level and the pain/temperature level is a direct, exam-visible consequence of where the spinothalamic tract crosses relative to dorsal column and corticospinal anatomy.

Example 2 — Interpreting an opioid dose-response and estimating relative potency. A hospital pain service is comparing intravenous morphine and hydromorphone. Published equianalgesic tables state:

A patient has been receiving morphine 4 mg IV every 3 hours (32 mg/24 h) with adequate but not complete analgesia, and the team wants to switch to hydromorphone. Reasoning: The equianalgesic ratio is 10 mg morphine ≈ 1.5 mg hydromorphone, i.e., hydromorphone is roughly 10/1.5 ≈ 6.7-fold more potent by mass on a μ-receptor-mediated analgesic basis (higher affinity/efficacy per milligram, not a difference in mechanism — both are μ-opioid agonists). Converting the current 24-hour morphine dose: 32 mg morphine × (1.5 mg hydromorphone / 10 mg morphine) = 4.8 mg hydromorphone per 24 hours. Because of incomplete cross-tolerance between different opioids at the receptor level (a well-documented clinical phenomenon), standard practice reduces the calculated equianalgesic dose by 25–50% when switching drugs, giving a starting dose of roughly 2.4–3.6 mg/24 h, then titrating upward based on response. This calculation is a direct clinical application of receptor pharmacology: because both drugs act at the same μ-opioid receptor pool in the PAG/RVM/dorsal horn circuit described above, their effects are quantitatively interconvertible via a potency ratio, but incomplete cross-tolerance reflects receptor-level adaptations (e.g., differential desensitization) that are not identical for every agonist.

Patrick Wall (1925–2001) was a British neurophysiologist trained at Oxford and Yale who spent much of his career investigating spinal cord physiology, first at MIT and later at University College London. Ronald Melzack (1929–2019), a Canadian psychologist trained under Donald Hebb at McGill, brought a background in the psychology of pain, having documented how amputees experience "phantom limb" pain that could not be explained by any peripheral injury at all — a phenomenon that convinced him pain must involve substantial central processing rather than a simple peripheral-to-cortex relay. Their collaboration, beginning when Melzack visited Wall's laboratory, combined Wall's electrophysiological rigor with Melzack's clinical and psychological insight into cases — chronic pain, phantom limb pain, causalgia — that specificity theory could not explain.

Understanding ascending pain pathways and descending modulation underlies nearly all modern pain medicine. Spinal cord stimulators and TENS units are direct clinical descendants of gate control theory, using large-fiber electrical stimulation to close the dorsal horn gate in patients with chronic neuropathic pain, failed back surgery syndrome, and complex regional pain syndrome. Epidural and intrathecal opioid administration exploits the dense μ-receptor population in the dorsal horn itself, allowing potent analgesia (e.g., during labor or major surgery) at doses far lower than systemic administration, minimizing side effects like respiratory depression. Multimodal analgesia protocols in surgery deliberately combine drugs acting at different nodes of this circuit — local anesthetics blocking peripheral conduction, NSAIDs reducing peripheral sensitization, gabapentinoids damping dorsal horn hyperexcitability, and opioids engaging the descending PAG–RVM system — reducing total opioid exposure and thereby addiction risk, a major public health priority given the ongoing opioid epidemic. Understanding incomplete cross-tolerance and receptor pharmacology (Worked Example 2) is now routine practice in palliative care and anesthesiology dosing calculators. Research into RVM "on-cell" facilitation is reshaping how clinicians think about chronic pain and opioid-induced hyperalgesia, where prolonged opioid use can paradoxically sensitize descending facilitatory circuits. Finally, the affective-motivational limb of the spinothalamic system (cingulate, insula) is a target for emerging non-opioid approaches, including mindfulness-based and cognitive interventions that modulate the same descending circuitry pharmacologically engaged by placebo and naloxone-reversible stimulation — and this same thalamocortical machinery sets up the next question in sensory neuroscience: how somatosensory information, pain included, is spatially organized into Sensory Maps and Cortical Topography.

Ask the tutor: How exactly do RVM "on-cells" and "off-cells" get identified experimentally, and what triggers the switch toward chronic pain facilitation?