scholarly journals Information about contact force and surface texture is mixed in the firing rates of cutaneous afferent neurons

2020 ◽  
Author(s):  
Monica Liu ◽  
Aaron Paul Batista ◽  
Sliman J Bensmaia ◽  
Douglas John Weber
Keyword(s):  
Nerve Fibers ◽  
Contact Forces ◽  
Sample Population ◽  
Textured Surfaces ◽  
Drg Neurons ◽  
Firing Rates ◽  
Haptic Interactions ◽  
Object Properties ◽  
Tactile Stimuli ◽  
Object Features

Tactile nerve fibers convey information about many features of haptic interactions, including the force and speed of contact, as well as the texture and shape of the objects being handled. How we perceive these object features is relatively unaffected by the forces and movements we use when interacting with the object. Since signals related to contact events and object properties are mixed in the responses of tactile fibers, our ability to disentangle these different components of our tactile experience implies that they are demultiplexed as they propagate along the neuraxis. To understand how texture and contact mechanics are encoded together by tactile fibers, we studied the activity of multiple neurons recorded simultaneously in the cervical dorsal root ganglia (DRG) of two anesthetized Rhesus monkeys while textured surfaces were applied to the glabrous skin of the fingers and palm using a handheld probe. A transducer at the tip of the textured probe measured contact forces as tactile stimuli were applied at different locations on the finger-pads and palm. We examined how a sample population of DRG neurons encode force and texture and found that firing rates of individual neurons are modulated by both force and texture. In particular, slowly-adapting (SA) neurons were more responsive to force than texture, and rapidly-adapting (RA) neurons were more responsive to texture than force. While force could be decoded accurately throughout the entire contact interval, texture signals were most salient during onset and offset phases of the contact interval.

scholarly journals Neurophysiology of Prehension. III. Representation of Object Features in Posterior Parietal Cortex of the Macaque Monkey

2007 ◽  
Vol 98 (6) ◽  
pp. 3708-3730 ◽  
Author(s):  
Esther P. Gardner ◽  
K. Srinivasa Babu ◽  
Soumya Ghosh ◽  
Adam Sherwood ◽  
Jessie Chen
Keyword(s):  
Parietal Cortex ◽  
Posterior Parietal Cortex ◽  
Macaque Monkey ◽  
Firing Rates ◽  
Video Recordings ◽  
Object Properties ◽  
Area 5 ◽  
Somatosensory Input ◽  
Posterior Parietal ◽  
Object Features

Neurons in posterior parietal cortex (PPC) may serve both proprioceptive and exteroceptive functions during prehension, signaling hand actions and object properties. To assess these roles, we used digital video recordings to analyze responses of 83 hand-manipulation neurons in area 5 as monkeys grasped and lifted objects that differed in shape (round and rectangular), size (large and small spheres), and location (identical rectangular blocks placed lateral and medial to the shoulder). The task contained seven stages—approach, contact, grasp, lift, hold, lower, relax—plus a pretrial interval. The four test objects evoked similar spike trains and mean rate profiles that rose significantly above baseline from approach through lift, with peak activity at contact. Although representation by the spike train of specific hand actions was stronger than distinctions between grasped objects, 34% of these neurons showed statistically significant effects of object properties or hand postures on firing rates. Somatosensory input from the hand played an important role as firing rates diverged most prominently on contact as grasp was secured. The small sphere—grasped with the most flexed hand posture—evoked the highest firing rates in 43% of the population. Twenty-one percent distinguished spheres that differed in size and weight, and 14% discriminated spheres from rectangular blocks. Location in the workspace modulated response amplitude as objects placed across the midline evoked higher firing rates than positions lateral to the shoulder. We conclude that area 5 neurons, like those in area AIP, integrate object features, hand actions, and grasp postures during prehension.

Encoding timing and intensity in the ventral cochlear nucleus of the cat

1986 ◽  
Vol 56 (2) ◽  
pp. 261-286 ◽  
Author(s):  
W. S. Rhode ◽  
P. H. Smith
Keyword(s):  
Firing Rate ◽  
Cochlear Nucleus ◽  
Auditory Nerve ◽  
Cell Types ◽  
Nerve Fibers ◽  
Frequency Selectivity ◽  
Firing Rates ◽  
Ventral Cochlear Nucleus ◽  
Auditory Nerve Fibers ◽  
Different Cell Types

Physiological response properties of neurons in the ventral cochlear nucleus have a variety of features that are substantially different from the stereotypical auditory nerve responses that serve as the principal source of activation for these neurons. These emergent features are the result of the varying distribution of auditory nerve inputs on the soma and dendrites of the various cell types within the nucleus; the intrinsic membrane characteristics of the various cell types causing different responses to the same input in different cell types; and secondary excitatory and inhibitory inputs to different cell types. Well-isolated units were recorded with high-impedance glass microelectrodes, both intracellularly and extracellularly. Units were characterized by their temporal response to short tones, rate vs. intensity relation, and response areas. The principal response patterns were onset, chopper, and primary-like. Onset units are characterized by a well-timed first spike in response to tones at the characteristic frequency. For frequencies less than 1 kHz, onset units can entrain to the stimulus frequency with greater precision than their auditory nerve inputs. This implies that onset units receive converging inputs from a number of auditory nerve fibers. Onset units are divided into three subcategories, OC, OL, and OI. OC units have extraordinarily wide dynamic ranges and low-frequency selectivity. Some are capable of sustaining firing rates of 800 spikes/s at high intensities. They have the smallest standard deviation and coefficient of variation of the first spike latency of any cells in the cochlear nuclei. OC units are candidates for encoding intensity. OI and OL units differ from OC units in that they have dynamic ranges and frequency selectivity ranges much like those of auditory nerve fibers. They differ from one another in their steady-state firing rates; OI units fire mainly at the onset of a tone. OI units also differ from OL units in that they prefer frequency sweeps in the low to high direction. Primary-like-with-notch (PLN) units also respond to tones with a well-timed first spike. They differ from onset cells in that the onset peak is not always as precise as the spontaneous rate is higher. A comparison of spontaneous firing rate and saturation firing rate of PLN units with auditory nerve fibers suggest that PLN units receive one to four auditory nerve fiber inputs. Chopper units fire in a sustained regular manner when they are excited by sound.(ABSTRACT TRUNCATED AT 400 WORDS)

scholarly journals Sensory computations in the cuneate nucleus of macaques

2021 ◽  
Vol 118 (49) ◽  
pp. e2115772118
Author(s):  
Aneesha K. Suresh ◽  
Charles M. Greenspon ◽  
Qinpu He ◽  
Joshua M. Rosenow ◽  
Lee E. Miller ◽  
...  
Keyword(s):  
Cortical Neurons ◽  
Receptive Fields ◽  
Nerve Fibers ◽  
Cuneate Nucleus ◽  
Local Skin ◽  
Response Properties ◽  
Conventional View ◽  
Random Dot Patterns ◽  
Object Features ◽  
Spatiotemporal Response

Tactile nerve fibers fall into a few classes that can be readily distinguished based on their spatiotemporal response properties. Because nerve fibers reflect local skin deformations, they individually carry ambiguous signals about object features. In contrast, cortical neurons exhibit heterogeneous response properties that reflect computations applied to convergent input from multiple classes of afferents, which confer to them a selectivity for behaviorally relevant features of objects. The conventional view is that these complex response properties arise within the cortex itself, implying that sensory signals are not processed to any significant extent in the two intervening structures—the cuneate nucleus (CN) and the thalamus. To test this hypothesis, we recorded the responses evoked in the CN to a battery of stimuli that have been extensively used to characterize tactile coding in both the periphery and cortex, including skin indentations, vibrations, random dot patterns, and scanned edges. We found that CN responses are more similar to their cortical counterparts than they are to their inputs: CN neurons receive input from multiple classes of nerve fibers, they have spatially complex receptive fields, and they exhibit selectivity for object features. Contrary to consensus, then, the CN plays a key role in processing tactile information.

Response Patterns in Second Somatosensory Cortex (SII) of Awake Monkeys to Passively Applied Tactile Gratings

2000 ◽  
Vol 84 (2) ◽  
pp. 780-797 ◽  
Author(s):  
J. R. Pruett ◽  
R. J. Sinclair ◽  
H. Burton
Keyword(s):  
Somatosensory Cortex ◽  
Hand Movement ◽  
Scanning Speed ◽  
Random Force ◽  
Groove Width ◽  
Hand Position ◽  
Sample Population ◽  
Firing Rates ◽  
Passive Touch ◽  
Movement Parameters

This experiment explored the effects of controlled manipulations of three parameters of tactile gratings, groove width (1.07–2.53 mm), contact force (30–90 g), and scanning speed (40–120 mm/s), on the responses of cells in second somatosensory cortex (SII) of awake monkeys that were performing a groove-width classification task with passively presented stimuli. A previous experiment involving an active touch paradigm demonstrated that macaque SII cells code groove-width and hand-movement parameters in their average firing rates. The present study used a passive-touch protocol to remove somatosensory activation related to hand movements that accompany haptic exploration of surfaces. Monkeys maintained a constant hand position while a robotic device delivered stimulation with tactile gratings to a single stabilized finger pad. Single-unit recordings isolated 216 neurons that were retrospectively assigned to SII on histological criteria. Firing patterns for 86 of these SII cells were characterized in detail, while monkeys classified gratings as rough (1.90 and 2.53 mm groove widths) or smooth (1.07 and 1.42 mm groove widths), with trial-wise random, parametric manipulation of force or speed; the monkeys compared 1.07 versus 1.90 mm and 1.42 versus 2.53 mm in alternating blocks of trials. We studied 33 cells with systematic variation of groove width and force, 49 with groove width and speed, and four with all three variables. Sixty-three cells were sensitive to groove width, 43 to force (effects of random force in speed experiments contributed to N), and 34 to speed. Relatively equal numbers of cells changed mean firing rates as positive or negative functions of increasing groove width, force, and/or speed. Cells typically changed mean firing rates for two or three of the independent variables. Effects of groove width, force, and speed were additive or interactive. The variety of response functions was similar to that found in a prior study of primary somatosensory cortex (SI) that used passive touch. The SII sample population showed correlated changes (both positive and negative) in firing rates with increasing groove width and force and to a lesser degree, with increasing groove width and speed. This correlation is consistent with human psychophysical studies that found increasing groove width and force increase perceived roughness magnitude, and it strengthens the argument for SII's direct involvement in roughness perception.

scholarly journals Ca2+ toxicity due to reverse Na+/Ca2+ exchange contributes to degeneration of neurites of DRG neurons induced by a neuropathy-associated Nav1.7 mutation

2015 ◽  
Vol 114 (3) ◽  
pp. 1554-1564 ◽  
Author(s):  
M. Estacion ◽  
B. P. S Vohra ◽  
S. Liu ◽  
J. Hoeijmakers ◽  
C. G. Faber ◽  
...  
Keyword(s):  
Nerve Fibers ◽  
Small Fiber Neuropathy ◽  
Missense Mutations ◽  
Sodium Calcium Exchanger ◽  
Drg Neurons ◽  
Persistent Currents ◽  
Burning Pain ◽  
Reverse Mode ◽  
Small Fiber ◽  
High K

Gain-of-function missense mutations in voltage-gated sodium channel Nav1.7 have been linked to small-fiber neuropathy, which is characterized by burning pain, dysautonomia and a loss of intraepidermal nerve fibers. However, the mechanistic cascades linking Nav1.7 mutations to axonal degeneration are incompletely understood. The G856D mutation in Nav1.7 produces robust changes in channel biophysical properties, including hyperpolarized activation, depolarized inactivation, and enhanced ramp and persistent currents, which contribute to the hyperexcitability exhibited by neurons containing Nav1.8. We report here that cell bodies and neurites of dorsal root ganglion (DRG) neurons transfected with G856D display increased levels of intracellular Na+ concentration ([Na+]) and intracellular [Ca2+] following stimulation with high [K+] compared with wild-type (WT) Nav1.7-expressing neurons. Blockade of reverse mode of the sodium/calcium exchanger (NCX) or of sodium channels attenuates [Ca2+] transients evoked by high [K+] in G856D-expressing DRG cell bodies and neurites. We also show that treatment of WT or G856D-expressing neurites with high [K+] or 2-deoxyglucose (2-DG) does not elicit degeneration of these neurites, but that high [K+] and 2-DG in combination evokes degeneration of G856D neurites but not WT neurites. Our results also demonstrate that 0 Ca2+ or blockade of reverse mode of NCX protects G856D-expressing neurites from degeneration when exposed to high [K+] and 2-DG. These results point to [Na+] overload in DRG neurons expressing mutant G856D Nav1.7, which triggers reverse mode of NCX and contributes to Ca2+ toxicity, and suggest subtype-specific blockade of Nav1.7 or inhibition of reverse NCX as strategies that might slow or prevent axon degeneration in small-fiber neuropathy.

scholarly journals Electroacupuncture Reduces Carrageenan- and CFA-Induced Inflammatory Pain Accompanied by Changing the Expression of Nav1.7 and Nav1.8, rather than Nav1.9, in Mice Dorsal Root Ganglia

2013 ◽  
Vol 2013 ◽  
pp. 1-8 ◽  
Author(s):  
Chun-Ping Huang ◽  
Hsiang-Ni Chen ◽  
Hong-Lin Su ◽  
Ching-Liang Hsieh ◽  
Wei-Hsin Chen ◽  
...  
Keyword(s):  
Sodium Channels ◽  
Dorsal Root Ganglia ◽  
Inflammatory Pain ◽  
Dorsal Root ◽  
Low Frequency ◽  
Thermal Hyperalgesia ◽  
Nerve Fibers ◽  
Drg Neurons ◽  
Plantar Surface ◽  
Voltage Gated Sodium Channels

Several voltage-gated sodium channels (Navs) from nociceptive nerve fibers have been identified as important effectors in pain signaling. The objective of this study is to investigate the electroacupuncture (EA) analgesia mechanism by changing the expression of Navs in mice dorsal root ganglia (DRG). We injected carrageenan and complete Freund's adjuvant (CFA) into the mice plantar surface of the hind paw to induce inflammation and examined the antinociception effect of EA at the Zusanli (ST36) acupoint at 2 Hz low frequency. Mechanical hyperalgesia was evaluated by using electronic von Frey filaments, and thermal hyperalgesia was assessed using Hargreaves' test. Furthermore, we observed the expression and quality of Navs in DRG neurons. Our results showed that EA reduced mechanical and thermal pain in inflammatory animal model. The expression of Nav1.7 and Nav1.8 was increased after 4 days of carrageenan- and CFA-elicited inflammatory pain and further attenuated by 2 Hz EA stimulation. The attenuation cannot be observed in Nav1.9 sodium channels. We demonstrated that EA at Zusanli (ST36) acupoint at 2 Hz low-frequency stimulation attenuated inflammatory pain accompanied by decreasing the expression of Nav1.7 and 1.8, rather than Nav1.9, sodium channels in peripheral DRG neurons.

Chloride Channels in Nociceptors

2019 ◽  
pp. 345-363
Author(s):  
Uhtaek Oh ◽  
Jooyoung Jung
Keyword(s):  
Transient Receptor Potential ◽  
Trp Channels ◽  
Transient Receptor Potential Channels ◽  
Primary Afferent Depolarization ◽  
Anion Channels ◽  
Drg Neurons ◽  
Anoctamin 1 ◽  
Central Process ◽  
Primary Afferent ◽  
Tactile Stimuli

Pain may be induced by activation of various ion channels expressed in primary afferent neurons. These channels function as molecular sensors that detect noxious chemical, temperature, or tactile stimuli and transduce them into nociceptor electrical signals. Transient receptor potential channels are good examples because they are activated by chemicals, heat, cold, and acid in nociceptors. Anion channels were little studied in nociception because of the notion that anion channels might induce hyperpolarization of nociceptors on opening. In contrast, opening of Cl- channels in dorsal root ganglion (DRG) neurons depolarizes sensory neurons, resulting in excitation of nociceptors, thereby inducing pain. Anoctamin 1(ANO1)/TMEM16A is a Ca2+-activated Cl- channel expressed mainly in small DRG neurons, suggesting a nociception role. ANO1 is a heat sensor that detects heat over 44°C. Ano1-deficient mice elicit less nocifensive behaviors to hot temperatures. In addition, mechanical allodynia and hyperalgesia induced by inflammation or nerve injury are alleviated in Ano1 -/- mice. More important, Ano1 transcripts are increased in chronic pain models. Bestrophin 1 (Best1) is another Ca2+-activated Cl- channel expressed in nociceptors. Best1 is increased in axotomized DRG neurons. The role of Best1 in nociception is not clear. GABAA receptors are in the central process of DRG neurons; GABA depolarizes the primary afferents. This depolarization consists of primary afferent depolarization essential for inhibiting nociceptive input to second-order neurons in the spinal cord, regulating pain signals to the brain. Thus, although Cl- channels in nociceptors are not as numerous as TRP channels, their role in nociception is distinct and significant.

Objective classification of motion- and direction-sensitive neurons in primary somatosensory cortex of awake monkeys

1986 ◽  
Vol 56 (3) ◽  
pp. 598-622 ◽  
Author(s):  
S. Warren ◽  
H. A. Hamalainen ◽  
E. P. Gardner
Keyword(s):  
Somatosensory Cortex ◽  
Cortical Neurons ◽  
Statistical Tests ◽  
Direction Selectivity ◽  
Primary Somatosensory Cortex ◽  
Relative Distribution ◽  
Textured Surfaces ◽  
Simple Method ◽  
Firing Rates ◽  
Moving Stimuli

In order to classify movement-sensitive neurons in SI cortex, and to estimate their relative distribution, we have developed a new simple method for controlled motion of textured surfaces across the skin, as well as a set of objective criteria for determining direction selectivity. Moving stimuli were generated using 5 mm thick precision gear wheels, whose teeth formed a grafting. They were mounted on the shafts of low-torque potentiometers (to measure the speed and direction of movement) and rolled manually across the skin using the potentiometer shaft as an axle. As the grafting wheel was advanced, its ridges sequentially contacted a specific set of points on the skin, leaving gaps of defined spacing that were unstimulated. This stimulus was reproducible from trial to trial and produced little distention of the skin. Three objective criteria were used to categorize responses: the ratio of responses to motion in the most and least preferred directions [direction index (DI)], the difference between mean firing rates in the two directions divided by the average standard deviation [index of discriminability (delta'e)], and statistical tests. Neurons were classified as direction sensitive if DI greater than 35, delta's greater than or equal to 1.35 (equivalent to 75% correct discrimination by an unbiased observer), and firing rates in most- and least-preferred directions were significantly different (P less than 0.05). Good agreement was found between the three classification schemes. Recordings were made from 1,020 cortical neurons in the hand and forearm regions of primary somatosensory cortex (areas 3b, 1 and 2) of five macaque monkeys. Tangential motion across the skin was found to be an extremely effective stimulus for SI cortical neurons. Two hundred eighty six of 757 tactile neurons (38%) responded more vigorously to moving stimuli than to pressure or tapping the skin. One hundred twenty-one cells were tested with moving gratings and were classified according to their ability to differentiate movement in longitudinal and transverse directions. Responses to the moving gratings resembled those observed when stroking the skin with brushed, edges, or blunt probes. Three major types of firing patterns were found: motion sensitive, direction sensitive, and orientation sensitive. Motion-sensitive neurons (37%) responded to movement in both longitudinal and transverse directions with only slight difference in firing rates and interval distributions. Responses throughout the field were fairly uniform, and no clear point of maximum sensitivity was apparent. Direction-sensitive neurons (60%) displayed clear preferences for movement in one or more directions.4

Protein kinase G signaling pathway is involved in sympathetically maintained pain by modulating ATP-sensitive potassium channels

2021 ◽  
pp. rapm-2021-102539
Author(s):  
Huiming Li ◽  
Mengjuan Shang ◽  
Ling Liu ◽  
Xiaoyu Lin ◽  
Junfeng Hu ◽  
...  
Keyword(s):  
Protein Kinase ◽  
Signaling Pathway ◽  
Small Interfering Rna ◽  
Small Diameter ◽  
Nerve Fibers ◽  
Protein Kinase G ◽  
Intractable Pain ◽  
Drg Neurons ◽  
Sympathetically Maintained Pain ◽  
Interfering Rna

BackgroundSympathetically maintained pain (SMP) involves an increased excitability of dorsal root ganglion (DRG) neurons to sympathetic nerve stimulation and circulating norepinephrine. The current treatment of SMP has limited efficacy, and hence more mechanistic insights into this intractable pain condition are urgently needed.MethodsA caudal trunk transection (CTT) model of neuropathic pain was established in mice.Immunofluorescence staining, small interfering RNA, pharmacological and electrophysiological studies were conducted to test the hypothesis that norepinephrine increases the excitability of small-diameter DRG neurons from CTT mice through the activation of cyclic guanosine monophosphate-protein kinase G (cGMP-PKG) signaling pathway.ResultsBehavior study showed that CTT mice developed mechanical and heat hypersensitivities, which were attenuated by intraperitoneal injection of guanethidine. CTT mice also showed an abnormal sprouting of tyrosine hydroxylase-positive nerve fibers in DRG, and an increased excitability of small-diameter DRG neurons to norepinephrine, suggesting that CTT is a useful model to study SMP. Importantly, inhibiting cGMP-PKG pathway with small interfering RNA and KT5823 attenuated the increased sympathetic sensitivity in CTT mice. In contrast, cGMP activators (Sp-cGMP, 8-Br-cGMP) further increased sympathetic sensitivity. Furthermore, phosphorylation of ATP-sensitive potassium channel, which is a downstream target of PKG, may contribute to the adrenergic modulation of DRG neuron excitability.ConclusionsOur findings suggest an important role of cGMP-PKG signaling pathway in the increased excitability of small-diameter DRG neurons to norepinephrine after CTT, which involves an inhibition of the ATP-sensitive potassium currents through PKG-induced phosphorylation. Accordingly, drugs targeting this pathway may help to treat SMP.

Norepinephrine reduces ω-conotoxin-sensitive Ca2+ currents in renal afferent neurons in rats

2012 ◽  
Vol 302 (3) ◽  
pp. F350-F357 ◽  
Author(s):  
Tilmann Ditting ◽  
Peter Linz ◽  
Wolfgang Freisinger ◽  
Sonja Heinlein ◽  
Peter W. Reeh ◽  
...  
Keyword(s):  
Confocal Microscopy ◽  
Sympathetic Nerve ◽  
Nerve Fibers ◽  
Afferent Nerve ◽  
Nerve Endings ◽  
Renal Nerves ◽  
Drg Neurons ◽  
Voltage Gated ◽  
Adrenoceptor Antagonist ◽  
Afferent Nerve Fibers

Sympathetic efferent and peptidergic afferent renal nerves likely influence hypertensive and inflammatory kidney disease. Our recent investigation with confocal microscopy revealed that in the kidney sympathetic nerve endings are colocalized with afferent nerve fibers (Ditting T, Tiegs G, Rodionova K, Reeh PW, Neuhuber W, Freisinger W, Veelken R. Am J Physiol Renal Physiol 297: F1427–F1434, 2009; Veelken R, Vogel EM, Hilgers K, Amman K, Hartner A, Sass G, Neuhuber W, Tiegs G. J Am Soc Nephrol 19: 1371–1378, 2008). However, it is not known whether renal afferent nerves are influenced by sympathetic nerve activity. We tested the hypothesis that norepinephrine (NE) influences voltage-gated Ca2+ channel currents in cultured renal dorsal root ganglion (DRG) neurons, i.e., the first-order neuron of the renal afferent pathway. DRG neurons (T11–L2) retrogradely labeled from the kidney and subsequently cultured, were investigated by whole-cell patch clamp. Voltage-gated calcium channels (VGCC) were investigated by voltage ramps (−100 to +80 mV, 300 ms, every 20 s). NE and appropriate adrenergic receptor antagonists were administered by microperfusion. NE (20 μM) reduced VGCC-mediated currents by 10.4 ± 3.0% ( P < 0.01). This reduction was abolished by the α-adrenoreceptor inhibitor phentolamine and the α2-adrenoceptor antagonist yohimbine. The β-adrenoreceptor antagonist propranolol and the α1-adrenoceptor antagonist prazosin had no effect. The inhibitory effect of NE was abolished when N-type currents were blocked by ω-conotoxin GVIA, but was unaffected by other specific Ca2+ channel inhibitors (ω-agatoxin IVA; nimodipine). Confocal microscopy revealed sympathetic innervation of DRGs and confirmed colocalization of afferent and efferent fibers within in the kidney. Hence NE released from intrarenal sympathetic nerve endings, or sympathetic fibers within the DRGs, or even circulating catecholamines, may influence the activity of peptidergic afferent nerve fibers through N-type Ca2+ channels via an α2-adrenoceptor-dependent mechanism. However, the exact site and the functional role of this interaction remains to be elucidated.

Sign in / Sign up

Export Citation Format

Share Document