Sickle cell disease (SCD) is due to a point mutation in the beta chain of hemoglobin that causes polymerization of deoxyhemoglobin that distorts the shape of erythrocytes, contributing to vascular obstruction and ischemia of tissues and organs [1
]. A hallmark feature of the disease is severe pain that arises during acute sickling events, as well as the recently recognized and less well understood chronic pain syndromes that develop in many of these individuals. Patients report heightened sensitivity to touch and spontaneous pain, suggesting complex physiological underpinnings that may include both inflammatory and neuropathic etiologies [2
]. The array of pain descriptors and triggers for acute and chronic pain suggest contribution from both central and
peripheral mechanisms, involving changes in neural signaling, gene expression and plasticity [3
]. Primary afferent sensitization, in particular, is a correlate of psychophysical measurements of hyperalgesia, characterized in part by increased responses to suprathreshold peripheral stimuli.
Recently, we reported that mice with severe sickle cell disease, which express only human sickle hemoglobin in circulating erythrocytes (HbSS mice) [4
], exhibit marked chronic mechanical hypersensitivity. Histological analysis of sickle mouse skin has revealed increased sensory innervation, elevated calcitonin gene-related peptide and substance P protein levels and diminished skin thickness [3
], all of which may contribute to the well-documented hyperalgesia exhibited in these animals. Primary afferent recordings from both Aδ-mechanoreceptor (AM) (high-threshold
4mN) and unmyelinated C fiber nociceptors showed enhanced mechanically-evoked action potential firing to suprathreshold force in HbSS mice compared to control HbAA mice that express 100% normal human hemoglobin [5
]. Yet, it is likely that these neural populations alone do not explain the complex mechanical allodynia phenotype reported in humans [6
], particularly because differences in mechanical firing rates between HbSS and controls were observed only at high mechanical forces in nociceptors [5
]. It is possible that sensitization of traditionally non-nociceptive Aβ or rapidly-adapting Aδ (D-hair) afferents may be involved, in addition to sensitization of CNS pathways. Because sickle mice were very sensitive to low intensity von Frey thresholds, we asked whether low threshold mechanoreceptors are sensitized.
Sickle mice exhibit mechanical allodynia
In humans, mechanical allodynia is likely a component of the complex pain associated with SCD [6
]. It is not known whether sickle mice exhibit a similar behavioral phenotype. Using our recently developed Light Touch Behavioral Assay [7
], we were able to measure allodynia-like behavioral responses to both punctate and dynamic light touch stimuli in sickle mice. We found that HbSS mice exhibited a 2-fold increase in paw withdrawal frequency to repeated application of a 0.7mN von Frey monofilament to the plantar hindpaw (Figure A, p
0.05). The second aspect of the assay recapitulated a dynamic mechanical stimulus, such as light stroke or wind, which is a correlate to increased pain in human SCD patients [6
]. Here, we gently stroked a “puffed” cotton swab across the plantar hindpaw skin, and recorded paw withdrawal frequency. HbSS mice exhibited an increased response (1.7-fold) to this dynamic touch (Figure B, p
0.001). Thus, by using two measurements, we were able to demonstrate a mechanical phenotype in sickle mice akin to mechanical allodynia in patients.
Figure 1 HbSS mice exhibit increased sensitivity to light-touch mechanical stimuli. Using the Light Touch Behavioral Assay, mechanical stimuli were applied to the glabrous skin of the hindpaws. The responses of both left and right hindpaws were counted and average (more ...)
Sickle mice exhibit increased mechanical responsiveness in light-touch primary afferents
While we previously found that nociceptors (myelinated and unmyelinated) in HbSS mice are sensitized to presumably noxious mechanical forces [5
], it is possible that light-touch mechanoreceptors are also sensitized and may contribute to mechanical allodynia in SCD. Therefore, we investigated the contribution of the cutaneous Aβ and D-hair afferents, which predominantly transmit non-nociceptive tactile sensation from the periphery. We quantified mechanically-evoked action potentials using the ex-vivo
saphenous skin-nerve preparation which innervates the hairy skin of the dorsal hindpaw, by recording from single cutaneous fibers and characterized the afferents by their conduction velocity and von Frey thresholds. We then applied increasing sustained force (5-200mN, 10
sec each) to each receptive field to measure firing to suprathreshold stimuli.
The mechanical thresholds for initial action potential responses did not differ for any fiber type in HbSS compared to HbAA control mice (Table
). In contrast, several fiber types from HbSS mice exhibited sensitization in the form of amplified action potential firing to suprathreshold stimuli. The greatest increase occurred in the rapidly adapting fiber subtypes that likely innervate hair follicles. Rapidly adapting Aβ fibers exhibited an average 75% increase in action potential firing across all force intensities (Figure G, p
0.05). At 200mN, RA-Aβ fibers from HbSS mice fired 3-fold more action potentials than HbAA controls. Additionally, there appears to be a small subpopulation of RA-Aβ fibers in HbSS mice that exhibit increased action potential firing at the onset of mechanical force, although we were unable to identify this small subgroup based on von Frey thresholds or conduction velocity for further electrophysiological testing (Figure C). Similarly, the rapidly adapting Aδ D-hair fibers also exhibited a 75% increase in overall firing across all forces (Figure H, p
0.05). On the other hand, the slowly adapting Aβ fibers, many of which innervate Merkel cells, showed a 25% increase in suprathreshold firing (Figure D, p
0.05). We further subtyped the SA-Aβ afferents into lower-threshold (VFT <4mN) and higher-threshold (VFT ≥4mN), because these fiber types exhibit different firing patterns to sustained force [8
] and because a small portion of slowly adapting Aβ fibers may be nociceptors [9
]. The enhanced mechanical firing in HbSS SA-Aβ afferents was restricted to the lower-threshold SA-Aβ afferents (Figure E, p
0.05). The higher-threshold afferents accounted for a small portion (25%) of total SA-Aβ fibers and exhibited no significant change (Figure F p
0.05). The conduction velocities of all Aβ fiber types did not differ between HbSS and HbAA controls. However, there was a slight, but significant decrease in conduction velocity in D-hair fibers in HbSS mice (Table
). Taken together, these data suggest that enhanced firing in light touch cutaneous afferents may contribute to the mechanical allodynia-type behavior in mice with sickle cell disease.
Summary of fiber properties in HbAA and HbSS mice
Figure 2 Mechanically-evoked action potential firing increases in HbSS mouse A-fibers. Using the skin-nerve preparation, all recordings were performed in the saphenous nerve and hairy skin of the dorsal hindpaw. Mechanical forces ranging 5-200mN (10sec) (more ...)
Overall locomotor activity does not correlate with mechanical allodynia in HbSS mice
To differentiate between an increased reflex response to light mechanical force, and an overall increase in locomotor activity or anxiety levels, we used an open field behavioral assay. HbSS mice exhibited decreased locomotor activity, traveling 65% less than HbAA controls (Figure A). We also quantified the amount of time spent in the center zone to measure anxiety-like behavior (Figure B). Neither HbAA nor HbSS mice avoided the center zone, indicating that anxiety-like behavior does not contribute to the mechanical hypersensitivity phenotype reported here and elsewhere [3
]. Additionally, no differences in exploratory behavior (Figure C) or immobile time (data not shown) were observed between genotypes. In sum, these data offer compelling evidence that the mechanical allodynia in HbSS mice is independent of other locomotor changes in sickle mice. Although speculative, this may indicate that the HbSS mice experience ongoing pain and move more slowly as a result, similar to the kinesiophobia reported in SCD human patients with heightened pain [10
This study broadens our understanding of the changes in the somatosensory system in sickle mice that contribute to their behavioral mechanical hypersensitivity. Here we show that several subtypes of low threshold mechanoreceptors that detect innocuous tactile information are sensitized to force in the form of enhanced suprathreshold firing. We and others [3
] have recently shown that sickle mice exhibit a heightened behavioral sensitivity to traditional von Frey filament threshold measurements, which may be in the noxious range. We also showed that Aδ- and C fiber-type nociceptors are sensitized to mechanical force in the form of enhanced suprathreshold firing [5
]. However, sensitized nociceptors, along with central sensitization of CNS pathways, may not fully account for the mechanical allodynia that is prevalent in humans because patients report enhanced sensitivity to wind currents and very light skin touch [6
]. Therefore, we further investigated the behavioral mechanical phenotype by using a punctate and dynamic light touch assay. Our data show that sickle mice are hypersensitive to very low threshold tactile stimuli. This behavioral phenotype is consistent with SCD-mediated allodynia in human patients and offers a new avenue to identify the cellular and molecular mechanisms that underlie it.
Somatosensory encoding of diverse tactile information from the physical environment is driven by input from a diverse array of mechanoreceptor neurons that are each tuned to detect specific qualities of the stimulus. Rapidly adapting Aβ fibers innervate guard hairs in hairy skin to detect dynamic stimuli such as wind currents, whereas Meissner’s corpuscles are found in the ridges of the glabrous skin where they detect low frequency vibration and microgeometric surface features such as corners and edges. The Aδ D-hair fibers innervate down or vellus hair follicles that are responsible for transmitting very light dynamic stimuli, including soft brush, stroke or light wind currents. Our finding that both types of myelinated hair follicle afferents (RA-Aβ and D-hair) from sickle mice were sensitized to force is interesting in light of the finding that sickle patients report enhanced sensitivity to wind currents, and increased hospitalizations of sickle patients are associated with elevated environmental wind speeds [6
]. It is possible that these subpopulations of hair follicle afferents are key detectors of environmental wind currents. Slowly adapting Aβ fibers innervate Merkel cells located at the epidermal-dermal border in both hairy and glabrous skin. Merkel cell afferents detect two-point discrimination, textures and patterns of object surfaces. The sensitization of all of these light touch afferent subtypes may contribute enhanced drive to the CNS that facilitates the mechanical hypersensitivity behavioral phenotype in sickle mice.
Importantly, sensitization in the spinal cord and higher brain centers also likely contributes to the behavioral allodynia in sickle mice and the tactile hypersensitivity in patients. Indeed, mechanical allodynia has been long attributed to sensitization of central mechanisms within the spinal cord and higher brain centers. Ongoing activity in nociceptive afferents after nerve injury has been shown to induce sensitization of second order neurons [12
] and supraspinal structures [13
], independent of putative increased Aβ branching in the spinal cord [14
]. The potential sensitization of tactile afferents, including Aβ afferents, after nerve injury or inflammation has largely been dismissed. However, recent evidence has shown that peripheral neuropathy increases both the sensitivity and prolongs the action potential discharge in myelinated Aβ neurons [15
]. Our findings are similar as they show that in the sickle cell model of chronic mechanical hypersensitivity, Aβ and Aδ tactile afferents exhibit increased action potential firing rates in response to intense mechanical force, regardless of the specific mechanisms that induce this enhanced firing. Importantly, these data highlight the rationale to investigate functional and expression changes in mechanoreceptor molecules expressed in low threshold afferent neurons during any injury or diseases that are associated with persistent or chronic mechanical pain in patients or pain-behavior in animal models.
Enhanced function of the Transient Receptor Potential Vanilloid 1 (TRPV1) channel underlies part of the behavioral hypersensitivity and mediates most of the C fiber nociceptor sensitization to intense force in sickle mice [5
]. However, since TRPV1 is not expressed in most non-nociceptive afferents, it is likely that other molecular mechanisms mediate sensitization of light touch myelinated afferents in sickle cell disease. One possibility is that other TRP channels members, such as TRPC1 (Transient Receptor Potential Cannonical 1), may contribute. TRPC1 is functionally expressed in these afferents [17
], is important to light touch [7
], and has modified channel partner proteins following inflammation [18
]. Alternatively, the acid-sensing ion channels (ASIC) have been implicated in mechanical sensitization [19
], are expressed in Aβ and Aδ fibers and may be sensitized by the inflammation associated with SCD. Interestingly, the recently described novel family of mechanically sensitive, pore-forming channels, Piezo 1 and 2 [20
] are plausible mechanotransduction candidates for contributing to the tactile allodynia observed in SCD. Future studies will be essential to determine the molecular mechanism(s) underlying sickle cell tactile allodynia and may open avenues to developing improved therapeutic strategies.