Microglia Density Decreases in the Rat Rostral Nucleus of the Solitary Tract

University of Nebraska at Omaha DigitalCommons@UNO

Psychology Faculty Publications Department of Psychology

7-4-2017 MICROGLIA DENSITY DECREASES IN THE RAT ROSTRAL NUCLEUS OF THE SOLITARY TRACT ACROSS DEVELOPMENT AND INCREASES IN AN AGE-DEPENDENT MANNER FOLLOWING DENERVATION Andrew J. Riquier University of Nebraska at Omaha

Suzanne I. Sollars University of Nebraska at Omaha, [email protected]

Follow this and additional works at: https://digitalcommons.unomaha.edu/psychfacpub Part of the Biological Psychology Commons

Recommended Citation Riquier, Andrew J. and Sollars, Suzanne I., "MICROGLIA DENSITY DECREASES IN THE RAT ROSTRAL NUCLEUS OF THE SOLITARY TRACT ACROSS DEVELOPMENT AND INCREASES IN AN AGE-DEPENDENT MANNER FOLLOWING DENERVATION" (2017). Psychology Faculty Publications. 186. https://digitalcommons.unomaha.edu/psychfacpub/186

This Article is brought to you for free and open access by the Department of Psychology at DigitalCommons@UNO. It has been accepted for inclusion in Psychology Faculty Publications by an authorized administrator of DigitalCommons@UNO. For more information, please contact [email protected]. Neuroscience 355 (2017) 36–48

MICROGLIA DENSITY DECREASES IN THE RAT ROSTRAL NUCLEUS OF THE SOLITARY TRACT ACROSS DEVELOPMENT AND INCREASES IN AN AGE-DEPENDENT MANNER FOLLOWING DENERVATION

ANDREW J. RIQUIER AND SUZANNE I. SOLLARS * Department of Psychology, University of Nebraska at Omaha, Omaha, NE 68182, USA INTRODUCTION Abstract—Microglia are critical for developmental pruning Neuroplasticity is noted in a variety of sensory systems and immune response to injury, and are implicated in facili- (e.g., vision: Hubel and Wiesel, 1970; Hoffmann and tating neural plasticity. The rodent gustatory system is Dumoulin, 2015; audition: Leake et al., 1999; Zhang highly plastic, particularly during development, and out- et al., 2002; somatosensation: Jacquin et al., 1993; comes following nerve injury are more severe in developing Omelian et al., 2016; gustation: Sollars et al., 2006). animals. The mechanisms underlying developmental plasticity in the taste system are largely unknown, making Lesion models indicate that effects of neural injury can dif- microglia an attractive candidate. To better elucidate micro- fer greatly depending upon the age at which the damage glia’s role in the taste system, we examined these cells in occurs (Towfighi et al., 1997; Howard et al., 2005; Sollars, the rostral nucleus of the solitary tract (rNTS) during normal 2005; Yager et al., 2006; Takahashi et al., 2009; Omelian development and following transection of the chorda tym- et al., 2016). Several mechanisms underlying develop- pani taste nerve (CTX). Rats aged 5, 10, 25, or 50 days mental differences in neuroplasticity have been sug- received unilateral CTX or no surgery and were sacrificed gested, such as neurotrophic factors (Monfils et al., four days later. Brain tissue was stained for Iba1 or CD68, 2006), differences in glucose metabolism (Pulsinelli and both the density and morphology of microglia were et al., 1982; Voorhies et al., 1986), NMDA neurotoxicity assessed on the intact and transected sides of the rNTS. (Sheng et al., 1994; Mitani et al., 1998; Manning et al., We found that the intact rNTS of neonatal rats (9–14 days) shows a high density of microglia, most of which appear 2008), myelination (Reeves et al., 2005), and chondroitin reactive. By 29 days of age, microglia density significantly sulfate proteoglycan levels (Milev et al., 1998; Matsui decreased to levels not significantly different from adults et al., 2005). Developmental variations in the immune and microglia morphology had matured, with most cells response may also play a role (Graeber et al., 1998; appearing ramified. CD68-negative microglia density Melzer et al., 2001; Bartel, 2012). increased following CTX and was most pronounced for juvenile and adult rats. Our results show that microglia density Microglia is highest during times of normal gustatory afferent pruning. Furthermore, the quantity of the microglia response is Microglia are central nervous system (CNS) immune cells higher in the mature system than in neonates. These find- derived from the early embryonic yolk sac that migrate ings link increased microglia presence with instances of into the brain prior to the closing of the blood–brain normal developmental and injury induced alterations in the barrier (Ginhoux et al., 2010). Microglia respond to dam- Ó rNTS. 2017 The Author(s). Published by Elsevier Ltd on age and pathogens, perform regular maintenance of behalf of IBRO. This is an open access article under the CC CNS homeostasis, and play a crucial role in both synaptic BY-NC-ND license (http://creativecommons.org/licenses/by- nc-nd/4.0/). and functional neuroplasticity both during development and following injury. Microglia pruning is activity and complement pathway dependent (Stevens et al., 2007; Key words: microglia, chorda tympani transection, gustatory Schafer et al., 2012), and results in both the removal of development, immune response to injury, nucleus of the excess synapses during development (Paolicelli et al., solitary tract, CD68. 2011) and neuron death due to loss of input following injury (Graeber et al., 1998). In the developing rodent *Corresponding author. Address: Department of Psychology, Univer- brain, microglia are known to vastly increase in number sity of Nebraska at Omaha, 419 Allwine Hall, Omaha, NE 68182, early in life, peaking approximately 2 weeks after birth USA. E-mail address: [email protected] (S. I. Sollars). (Alliot et al., 1999) and decreasing starting in the third Abbreviations: ANOVA, analysis of variance; CD68, cluster of postnatal week (Nikodemova et al., 2015). In the adult differentiation 68; CNS, central nervous system; CT, chorda tympani mammalian brain, microglia typically exhibit small cell nerve; CTX, chorda tympani nerve transection; Iba1, ionized calcium- bodies and long, thin, ramified branches (Wake et al., binding adapter molecule 1; NTS, nucleus of the solitary tract; P#, postnatal day #; PBS, phosphate-buffered saline; rNTS, rostral 2009; Bartel, 2012). These ramified microglia express a (gustatory) nucleus of the solitary tract; TBS, tris-buffered saline. multitude of receptors for various substances such as http://dx.doi.org/10.1016/j.neuroscience.2017.04.037 0306-4522/Ó 2017 The Author(s). Published by Elsevier Ltd on behalf of IBRO. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). 36 A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 37 cytokines, neurotransmitters, and hormones, as well as current study, the number of microglia were counted factors present during trauma and inflammation (reviewed and classified by morphology (reactive or ramified) in in Hanisch and Kettenmann, 2007; Kettenmann et al., the rNTS across development (9, 14, 30, or 55 days of 2011). Following immune challenge, ramified microglia age). This was done to determine if the time course of can be induced to change into activated microglia, which microglia presence in the developing rat rNTS is similar vary in appearance from bushy to ameba-like to rod-like, to that observed in other brain areas (Alliot et al., 1999; but typically have larger cell bodies, thicker processes Nikodemova et al., 2015), and to determine whether and fewer branches than when ramified (reviewed in microglia density differences correspond to known peri- Kettenmann et al., 2011). Activated microglia perform ods of CT terminal field pruning (Sollars et al., 2006; many functions including phagocytosis and secretion of Mangold and Hill, 2008). Microglia have been shown to cytotoxins and cytokines (Delgado et al., 1998; Webster increase in number and assume a reactive morphology et al., 2000). Microglia have been shown to increase in following adult CTX in mice (Bartel, 2012) but whether this number at the central terminals following peripheral nerve response differs based on surgical age or species is injury (Graeber et al., 1998; Bartel, 2012). The outcome of unknown. Examination of microglia following CTX in a the microglia response to injury can vary based on timing context where peripheral anatomical recovery is known relative to the injury (Rice et al., 2015) or developmental to occur (adulthood) and in which recovery is absent stage (Graeber et al., 1998; Melzer et al., 2001). (neonatal period) may implicate microglia as a mechanism influencing CTX-induced structural plasticity. To this Gustatory system plasticity end, we performed unilateral CTX on rats aged 5, 10, 25, or 50 days, quantified microglia in the rNTS, and classi- Peripheral and central morphology of the rodent gustatory fied them as morphologically reactive or ramified. system is highly malleable, rendering the system an excellent model for studying developmentally dependent neural changes and injury-induced structural plasticity EXPERIMENTAL PROCEDURES (Sollars et al., 2002; Sollars, 2005; Sollars et al., 2006; Mangold and Hill, 2008). Continual peripheral synaptic Animals plasticity is notable given the taste receptor cell turnover Thirty-eight female and male (19F/19M) Sprague–Dawley rate of approximately 10–14 days across all ages rats born at the University of Nebraska at Omaha vivarium (Hendricks et al., 2004; Perea-Martinez et al., 2013). were utilized. Day of birth was designated as postnatal The volume of central terminal gustatory fields in the day 0 (P0). To maintain developmental consistency, nucleus of the solitary tract (NTS) exhibit extensive prun- litters contained at least 6 pups and were culled to a ing throughout postnatal development (Sollars et al., maximum of 12. Rats were weaned at P25, maintained 2006; Mangold and Hill, 2008). One of the most exten- on Teklad chow ad libitum in clear Plexiglas tubs with sively studied of the gustatory nerves, the chorda tympani corncob bedding, and housed in conventional facilities (CT), is a branch of the facial nerve that transmits taste within a temperature controlled, 12-h light/dark cycle information from fungiform and foliate taste buds on the room. Each experimental group contained both sexes tongue to the NTS. The CT terminal field is located pri- and animals from at least two different litters. This study marily in the rostral-most one-third of the NTS (rNTS; was approved by and followed the standards of the May and Hill, 2006; Corson and Hill, 2011). CT terminal University of Nebraska Institutional Animal Care and field volume is enlarged at postnatal day 15 (P15) and Use Committee. P25, and decreases in volume until adulthood (Sollars et al., 2006; Mangold and Hill, 2008). Both central and peripheral gustatory anatomy is Microglia response to chorda tympani transection modified by transection of gustatory nerves or dietary (CTX) manipulation (St. John et al., 1995; Krimm and Hill, 1997; Sollars et al., 2002; Sollars, 2005; Sollars et al., Twenty-six rats received unilateral (right) CTX at either 5 2006; Mangold and Hill, 2007). Developmental differ- (Iba1, n = 4), 10 (Iba1, n = 4; CD68, n = 5) 25 (Iba1, ences in fungiform taste bud and papillae recovery are n = 4) or 50 (Iba1, n = 4; CD68, n = 5) days of age, markedly pronounced following CT transection (CTX). In corresponding to previously examined developmental adult (>P40) rats, CTX results in transient peripheral time points (Sollars, 2005). Unilateral CTX was performed and behavioral alterations (Spector and Grill, 1992; St. as described previously (Sollars et al., 2002; Sollars, John et al., 1995; St. John et al., 1997) that are recovered 2005; Martin and Sollars, 2015). Briefly, anesthetized following CT regeneration approximately 40 days post- (BrevitalÒ Sodium; 60 mg/kg i.p.) rats received a small surgery (St. John et al., 1995; Kopka et al., 2000). Follow- incision in the ventral portion of the neck and the right ing neonatal CTX (10 days of age) the CT does not CT was exposed and sectioned. The left CT remained regenerate, and peripheral alterations are more severe intact to serve as an internal control. Sham animals and permanent (Sollars et al., 2002; Sollars, 2005). underwent identical procedures but both the left and right Microglia are implicated in activity-dependent pruning CT remained intact. Animals were maintained on a heat- of synapses during normal development (Paolicelli et al., ing pad and monitored until recovery, which typically took 2011; Schafer et al., 2012), thus it seems likely they are less than 30 min. Animals were allowed to survive for involved in the postnatal pruning of the CT terminal field 4 days following surgery, based on previous research (Sollars et al., 2006; Mangold and Hill, 2008). In the showing that microglia numbers following peripheral injury 38 A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 peak at approximately this time (Graeber et al., 1998; CD68 An antibody against Cluster of Differentiation 68 Bartel, 2012). (CD68) is expressed by macrophages/monocytes (Suzuki et al., 2014) and has been used to identify phagocytic microglia (Zabel et al., 2016). Following two five-minute Developmental Progression of microglia in the rNTS rinses in 0.01 M tris-buffered saline (TBS; pH 7.6) con- Microglia do not change in number on the contralateral taining 0.025% Triton X-100, sections were immersed in side of the NTS following adult mouse CTX (Bartel, a blocking solution of 10% normal goat serum and 1% 2012), but this information is not known for rats at various bovine serum albumin in 0.01 M TBS for two hours at stages of development. Unilateral CTX has been shown room temperature. Tissue was incubated in rabbit poly- to induce alterations on the side contralateral to the injury clonal anti-CD68 (Abcam, Cambridge, MA; 1:500 diluted in regard to taste bud volume (Guagliardo and Hill, 2007; in 1% bovine serum albumin in 0.01 M TBS) overnight Li et al., 2015), as well as in electrophysiological nerve at 4 °C. Following a rinse cycle, sections were placed in activity (Wall and McCluskey, 2008; Martin and Sollars, 0.3% H2O2 in methanol for 30 min. 2015). To ensure the effect of surgery on microglia number was contained to the rNTS ipsilateral to the CTX, two Secondary antibody. All tissue was rinsed, then animals from each age condition were sacrificed at either incubated with biotinylated goat anti-rabbit IgG (Vector 9, 14, 29, or 54 days of age and processed as age- Laboratories, Burlingame, CA; 1:1000) diluted in 2% matched, non-surgical controls. Non-surgical animals normal goat serum and 0.3% Triton X-100 for 2 h at were also used to assess microglia numbers in the intact room temperature. After rinsing, tissue was placed in an rat rNTS across development. avidin–biotin complex (Vector Laboratories, Burlingame, CA) solution containing 0.1% Triton for 1 h. Following a subsequent rinse, sections were incubated in a Perfusion and tissue extraction peroxidase solution (0.01% H2O2, 0.05% Rats were overdosed with a mixture of a ketamine/ diaminobenzidine and 0.125% nickel ammonium xylazine (i.p.) and transcardially perfused with a sulfate). Sections were rinsed in dH2O, mounted on gel- modified Krebs solution (pH 7.3) followed by 4% coated slides, and air dried. paraformaldehyde. Brains were extracted and post-fixed in 4% paraformaldehyde for 1 h at room temperature Controls. For each brain processed for Iba1, the two and the brainstem was dissected from the cortex. ventral-most sections were treated as stated above with Brainstems were immersed in 30% sucrose (24 h) for the exception of excluding either the primary or cryoprotection, placed in embedding medium, flash secondary antibody. No stain was evident on Iba1 frozen, and stored at À80 °C until processing, similar to negative control sections (Fig. 1A–C). Spleen tissue previous work (Watson et al., 1986; Breza and Travers, was used as control tissue for CD68 staining due to the 2016). Prior to processing, brainstems were thawed over- high baseline numbers of CD68+ macrophages present night at 4 °C, then for 20 min at room temperature. Tissue (Olsson et al., 2011; company data sheet; Fig. 1D–F.). was further post-fixed for 24 h in room temperature 4% To minimize any potential effect of chromogen variability, paraformaldehyde, then horizontally sectioned at 40 mm each IHC run processed 2–4 brains simultaneously. on a vibrating microtome. Horizontal sections were used to facilitate comparisons to studies that have examined NTS identification and analysis by Dorsal–Ventral the CT terminal field following CTX (Sollars et al., 2006; zones Reddaway et al., 2012). To ensure consistency in identification of morphological landmarks, following processing and mounting, selected Immunohistochemistry sections were stained using the myelin stain hydrogen tetrachloroaurate (III) trihydrate (Sigma–Aldrich, St. Louis, MO, USA Schmued, 1990; Corson and Hill, Primary antibodies. Iba1 An antibody against ionized 2011). Following dH2O rinsing, slides were immersed in calcium-binding adapter molecule 1 (Iba1) was used for 0.2% hydrogen tetrachloroaurate (III) trihydrate dissolved microglia visualization. Iba1 is a protein expressed by in 0.02 M PBS (pH 7.4) for 4 h. Slides were rinsed, placed microglia (Imai et al., 1996), and is used for microglia in 2.5% sodium thiosulfate for 5 min, then immersed for identification (Bartel, 2012; Bartel and Finger, 2013). All 30 min in running dH2O. After drying, tissue was rinse cycles were three 10-min washes in 0.01 M immersed in xylenes for 5 min and coverslipped with phosphate-buffered saline (PBS; pH 7.4). All solutions DPX mounting medium. used this concentration of PBS as a diluent. After an initial Data were collected using a brightfield microscope rinse cycle, free-floating sections were incubated in 0.3% and Neurolucida software (MBF Bioscience) by a single, H2O2 in methanol for 30 min. Following rinses, sections trained, researcher blind to conditions. A grid of 250 mm were immersed in a blocking solution consisting of 2% X 250 mm squares was used to determine and trace the normal goat serum, 1% bovine serum albumin and 0.3% rNTS, defined as the rostral-most one-third of the entire Triton X-100 in PBS for 1 h at room temperature. Sections length of the NTS (Corson and Hill, 2011). The rNTS were incubated in rabbit polyclonal anti-Iba1 (Wako, Rich- was subdivided into contiguous dorsal, intermediate, mond, VA; 1:10,000 diluted in blocking solution) for 60 h and ventral zones (Fig. 2) each consisting of 4 sections, at 4 °C. similar to previous research (May and Hill, 2006; Sollars A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 39

Fig. 1. Immunohistochemical Controls. Images from the same adult rat brainstem of Iba1 control sections with (A) no primary antibody and (B) no secondary antibody. (C) Both primary and secondary antibody were included. CD68 control spleen tissue with (D) no primary antibody (E) no secondary antibody and (F) both primary and secondary antibodies. The light source was darkened equally on all images to facilitate visualization of the tissue. et al., 2006; Corson and Hill, 2011). The most prominent tive of microglia responsiveness (Avignone et al., 2015; landmarks are contained within the intermediate zone. Kato et al., 2016). In order to more accurately assess their Thus, the first area identified in each brain was the inter- role, all quantified Iba1+ microglia were classified as mediate zone, which is characterized by the visually dis- either ramified (soma diameter 8 mm, M = 7.61 mm tinct solitary tract passing in a rostral-to-caudal extent ± 0.02) or reactive (soma diameter > 8 mm, through the medial NTS. The facial nucleus in these sec- M = 13.11 mm ± 1.03; Fig. 3). These values were deter- tions was large and distinct. The dorsal zone was defined mined by averaging the diameter of a random, represen- as the sections starting 40 mm through 160 mm dorsal to tative sample of cells morphologically consistent with the intermediate zone. This area contained sections these classifications from previous research (Glenn where the fourth ventricle was at its largest medial–lateral et al., 1992; Bartel, 2012). extent. The facial nucleus was visible, though smaller than in the intermediate zone sections, and the solitary tract was limited to the lateral side of the NTS. The ventral Data analysis zone was defined as the sections starting 40 mm through 160 mm ventral to the intermediate zone. Relative to the For each experimental animal, microglia numbers were intermediate zone, the fourth ventricle was smaller or summed to obtain a single value per rNTS side. The not present and contained sections in which the NTS volume of each rNTS side was calculated by multiplying extended more rostrally than in the other zones. the total area by the section thickness (40 mm). To account for differences in NTS size across development, the density of microglia in the rNTS Microglia quantification and classification (microglia/200,000 mm3) was calculated by dividing total Cells were visualized using a 20Â (X 1.6) objective lens microglia numbers by the total volume of the rNTS in and were counted only if the cell body was entirely cubic mm for each animal, and multiplying that value by within the traced rNTS, similarly to the ‘‘counting box” 200,000. Densities represent standardized microglia method (Williams and Rakic, 1988). Analyses were con- quantities independent of total NTS volume, allowing for ducted on every other section (2 sections per zone, 6 sec- comparisons across age conditions. Density values tions per animal) to decrease the possibility of double were used for all statistical analyses except for counting cells. Counts were performed by placing overlay morphology. The percent of reactive microglia was markers in Neurolucida on the soma of each microglia, calculated using the total number of reactive microglia then analyzing marker counts using Neuroexplorer divided by the total number of microglia. One-way (MBF Bioscience). Microglia function can be broadly ANOVAs were conducted to determine if microglia determined by appearance, with enlarged somas sugges- density or the percent of reactive microglia in the intact 40 A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48

reactive microglia were tested for main effects of age and surgical side using an ANOVA. If the main effect of surgical side was significant, paired-samples t-tests were performed to compare transected and intact sides within each age group. All statistical tests were conducted in SPSS (version 23, IBM) using an alpha level of 0.05. P values for post hoc paired samples t- tests were corrected for false discovery rates (Benjamini and Hochberg, 1995).

Cell count correction The Abercrombie correction is used to account for overestimating counted cells (Abercrombie, 1946) using the equation Corrected Counts = Raw Counts * (section thickness/ [section thickness + cell height]) We randomly selected approximately 5% of the total number of microglia of reactive and ramiﬁed morphologies for each age and carefully measured soma diameter using Neurolucida. A correction utilizing the overall mean microglia diameter was performed for combined density analyses. Separate correction factors for each morphological classiﬁcation were used for morphology analyses.

RESULTS Developmental Progression of microglia in the rNTS

Microglia density. Microglia density and the percentage of reactive microglia were not significantly different between non-surgical rats, sham rats, and the intact rNTS of surgical animals for any age (all Ps > 0.10), consistent with previous work (Bartel, 2012). Therefore, data from non-surgical, sham, and the intact rNTS of surgical animals were pooled for analyses of normal development. Microglia density is greatest in the neonatal rNTS and decreases with age. There was a significant main effect of age on microglia density (F Fig. 2. (A) Schematic (not to scale) of the approximate location of the (3, 24) = 16.109, P < 0.001). Post-hoc tests revealed NTS in the rat. Gray lines indicate approximate subdivisions of the dorsal, intermediate, and ventral zones. Representative myelin that P9 and P14 animals had a significantly greater den- stained sections of the (B) dorsal (C) intermediate and (D) ventral sity of microglia than P29 (Ps < 0.05) and P54 zones of the NTS. The approximate location of the NTS is outlined in (Ps < 0.01) rats (Fig. 4A). The density of microglia was white. Arrows point to the solitary tract. VII: Facial Nucleus; 4 V: not significantly different when comparing animals aged Fourth Ventricle; R: Rostral; L: Lateral; V: Ventral. P9 and P14 (P > 0.10) or when comparing P29 to P54 rats (P > 0.10). These results suggest that microglia den- rNTS differed by age. Tukey post hoc tests were sity decreases across postnatal development, particularly performed to compare specific ages. between P14 and P29, at which point microglia density is A mixed-model ANOVA was conducted to determine not significantly different from adult (P54) levels. the effects of surgical age, surgical side, and sex on microglia density in the rNTS. If the main effects were Microglia morphology. The percent of microglia significant, paired-samples t-tests were used to assess exhibiting a reactive morphology in the rNTS decreases intact versus transected sides for each age group. This with age. There was a significant main effect of age on analysis was repeated for dorsal, intermediate, and the percent of reactive microglia in the rNTS (F(3, 24) ventral zones. The percent increase in microglia density = 8.287, P < 0.01; Figs. 4B and 5). P9 and P14 following CTX was calculated for each animal using the animals had 70% and 72% reactive microglia, and were following formula: [{(transected side density/intact side not significantly different from each other (P > 0.10), density) * 100} À 100]. The mean percent increase of indicating that the majority of microglia in the neonatal each age group was compared using a One-way rNTS are reactive. P9 and P14 animals had a ANOVA with Tukey post hoc tests. The percent of significantly larger percent of reactive microglia than A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 41

Microglia response to chorda tympani transection (CTX)

Overall microglia density. Microglia density is elevated in the transected-side rNTS relative to the intact-side rNTS four days after CTX (Fig. 6A). There was a significant main effect of surgery on microglia density (F(1, 8) = 131.46, P < 0.001). The density of Iba1+ microglia was significantly higher on the transected side of the rNTS Fig. 3. (A) Microglia in the adult brain are typically ramified, with small cell bodies and long, compared to the intact side for each branching processes. (B) Following nerve transection, microglia transition into a reactive morphology. Reactive microglia have larger cell bodies and shorter processes. surgical age: P5 (t(3) = 4.79, P = 0.017); P10 (t(3) = 4.60, P = 0.019); P25 (t(3) = 5.97, P = 0.009); and P50 (t(3) = 10.92, P = 0.002). These findings indicate that regardless of age, microglia density increases following CT axotomy. There was no significant main effect of sex (F(1, 8) = 0.49, P > 0.10) and no interactions with sex as a variable were significant (Ps > 0.10) suggesting that microglia responses to CTX are not different between males and females. Although there was no main effect of surgical age on microglia density (F(3, 8) = 2.43, P > 0.10), the interaction between surgical age and surgery was significant (F(3, 8) = 5.13, P < 0.05). This finding suggests that the effect of surgery differs based on the animal’s stage in development. The percent to which microglia density increased following CTX differed based on surgical age (F(3, 15) = 8.54, P < 0.01; Figs. 6B, 7). Animals that received CTX at 5 or 10 days of age displayed a significantly smaller microglia density increase compared to those that received CTX at 25 days of age (Ps < 0.05) or 50 (Ps < 0.05). The percentage of microglia density increase following P5 CTX was not significantly different from that of P10 CTX rats (P > 0.10). Similarly, rats transected at P25 and P50 had increases in microglia densities that were not significantly different from each other (Ps > 0.10). These results demonstrate that while microglia density increased four days after CTX in all ages tested, the magnitude of change was smaller in neonatal animals. Fig. 4. (A) Average microglia density (±SEM) and (B) the percent of reactive microglia (±SEM) in the intact rNTS across development. The percent of reactive microglia was calculated based on total Microglia morphology and function. Microglia assume microglia counts. Data were obtained from non-surgical animals, a reactive morphology following CTX. There was a * sham animals, and the intact rNTS of CTX rats. p < 0.05; significant main effect of surgical side (F(1, 24) = 31.48, **p < 0.01. P < . 001) as well as age (F(3, 24) = 3.091, P < 0.05) on the percent of microglia that displayed a reactive P29 and P54 rats (Ps < 0.01). The percent of reactive appearance. The interaction between surgical side and microglia in P29 and P54 animals were 23% and 25%, age trended toward significance (F(3, 24) = 2.71, values that are not significantly different (P > 0.10). P = 0.068), indicating that the impact of surgery on the This finding suggests that the gustatory NTS of juvenile percent of reactive microglia was not dependent on and adult rats primarily contains ramified microglia, surgical age (Fig. 6C). The percent of reactive microglia demonstrating that over the course of development did not differ based on side for P5 or P10 rats microglia morphology shifts from a predominantly (P > 0.10). The majority of microglia in the rNTS of reactive to ramified state. The decrease in microglia neonatal animals were classified as reactive, regardless density that occurs by P29 is therefore accompanied by of side (Fig. 7). The percent of microglia that were a shift in microglia morphology. reactive was significantly higher on the transected side 42 A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48

Fig. 5. Representative images of microglia in the intact rat rNTS at P9 (A), P14 (B), P29 (C), and P54 (D).

relative to controls for P25 (t(3) = 5.276, P = 0.016) and P50 (t(3) = 9.51, P = 0.002) animals. This finding demonstrates that the increase in microglia density observed on the transected side is accompanied by a shift in morphology from primarily ramified to mostly reactive microglia. The percent of reactive microglia on the transected side was not significantly different between P25 and P50 rats (P > 0.10), indicating that the morphological change induced by CTX reaches a mature state by P25. No CD68+ cells were detected in the rNTS of rats undergoing CTX at either 10 or 50 days of age (Fig. 8). CD68+ cells were present in spleen control tissue (Fig. 1) as well as in several locations throughout the neonatal brainstem (Fig. 8). The lack of CD68+ cells in the rNTS indicates that no phagocytic cells, including microglia, are present four days following CTX.

Microglia density by NTS zone. Although all surgical ages exhibited an increase in microglia density following CTX, the rNTS zones in which this increase occurred varied by age (Fig. 9). CTX increased microglia density throughout the rNTS in juvenile and adult rats, however the eﬀects of surgery were mostly limited to the ventral zone in neonates. Neonates. An increase in microglia density occurred in the dorsal (t(3) = 4.81, P = 0.017) and ventral (t(3) = 3.60, P = 0.037) zones of P5 animals, with a trend toward signiﬁcantly elevated microglia density in the intermediate Fig. 6. Developmental comparison of the increase and morphological change in the rostral nucleus of the solitary (rNTS) following transection of the chorda tympani nerve. (A) Average total zone (t(3) = 2.77, P = 0.07). The microglia density (±SEM), (B) the percent increase in microglia density (±SEM) between the increase in microglia density on the intact and transected rNTS, and (C) the percent of microglia with a reactive morphology (±SEM). transected side was limited to the A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 43

was elevated in the dorsal (t(3) = 4.426, P = 0.021), intermediate (t (3) = 5.72, P = 0.011), and ventral (t(3) = 4.50, P = 0.021) zones. P50 animals also exhibited an increase in microglia density in the dorsal (t(3) = 5.00, P = 0.015), intermediate (t (3) = 8.91, P = 0.003), and ventral (t(3) = 8.44, P = 0.003) zones.

DISCUSSION Our results show that microglia density in the rNTS of rats is higher in early postnatal ages (P9 and P14) than in juvenile (P29) and adult (P54) ages. The lower density noted at P29 remains consistent as the rats mature into adulthood. Microglia in the neonatal rNTS consists predominantly of reactive microglia, which mature by P29 into the ramified morphology typical of the adult brain. Following CTX, microglia density in the rNTS increases significantly on the side ipsilateral to the surgery, independent of whether CTX occurred in early postnatal or later ages. The ipsilateral rNTS in rats receiving surgery at either P25 or P50 not only shows an increase in microglia, but also the majority of those cells develop a reactive morphology. The morphological profile of microglia in P5 and P10 rats does not change as a result of the surgery. Microglia density increases to a greater degree on the CTX side at P25 and P50 than at P5 and P10. This finding may partially be explained by the normally elevated microglia density in neonatal rats relative to juvenile and adult conspecifics.

Microglia in the developing rNTS: Potential Insights regarding terminal field pruning Fig. 7. Microglia density and morphology in the intact and transected rostral nucleus of the solitary We found that microglia density in the tract. Iba1 + microglia phenotype in the intact (A, C, E, & G) and transected (B, D, F, & H) rNTS rat rNTS decreased from P14 to P29. four days after CTX occurring at P5 (A & B,) P10 (C & D), P25 (E & F), or P50 (G & H). Each surgical age represents the right and left rNTS of the same animal. Scale bar is 50 lm. Consistent with this finding, microglia numbers in the entire mouse brain peak two weeks after birth with a ventral zone of P10 animals (t(3) = 6.82, P = 0.006), subsequent 50% reduction from P14 with a trend toward increased microglia density in the to P42 due to a combination of apoptosis and reduced dorsal zone (t(3) = 2.73, P = 0.07) and no difference proliferation (Alliot et al., 1999; Nikodemova et al., between sides in the intermediate zone (t(3) = 2.29, 2015). We found a similar decrease between P14 and P > 0.10). P54, with a 50% reduction in microglia density. This Juveniles and Adults. In contrast to the neonatal developmental decrease in microglia density appears to conditions, microglia density increased in all zones of take place gradually, because only a 35% decrease from P25 and P50 animals. In P25 rats, microglia density neonatal density levels occurred between P14 and P29. 44 A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48

Fig. 8. There were no CD68+ cells present in the rostral nucleus of the solitary tract (rNTS) four days after CTX performed at (A) 10 or (B) 50 days of age. (C) shows CD68+ cells not in the rNTS, but elsewhere in the neonatal brainstem.

Our findings indicate that the percent of microglia exhibiting a reactive morphology in the intact rNTS decreases dramatically between P14 and P29, from 72% to 23%. The stark change in microglia morphology across development is consistent with what has been observed in several other brain regions (Schwarz et al., 2012). Reactive and activated microglia found in other neonatal brain regions are thought to be linked to synaptic remodeling (Dalmau et al., 1998; Fiske and Brunjes, 2000). Onset and progression of the largest period of CT terminal field pruning occurs prior to and around the time of weaning (P15/P25; Mangold and Hill, 2008) which corresponds to the highest densities of microglia we observed. Microglia have been implicated in postnatal synaptic pruning (Paolicelli et al., 2011; Schafer et al., 2012), and neonatal microglia express mRNA for factors associated with inflammation, apoptosis, and pruning, which are downregulated at adulthood (Crain et al., 2013). It is therefore possible that microglia are involved in developmental pruning of the CT terminal field.

Microglia response to CTX across development The current study demonstrates that while microglia increase in density four days after CTX in all ages examined, the degree and distribution of the increase varies developmentally. CTX elicits a greater relative increase in microglia Fig. 9. The microglia response in the rNTS to CTX at different ages, analyzed by zone. Average density in adults than in neonates. microglia density in the (A) dorsal, (B) intermediate, and (C) ventral zone of the intact and This may be attributed to a higher transected rNTS. *p < 0.05; **p < 0.01. baseline microglia density in the A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 45 intact rNTS of neonates relative to juvenile and adult Microglia response to adult CTX conspecifics. Although microglia density increased in all CTX in adult rats resulted in microglia assuming a rNTS zones following juvenile and adult CTX, the reactive morphology and an increase in microglia density increase after neonatal CTX was limited to the density four days after surgery. This increase was dorsal and ventral zones (P5) or ventral zone alone observed across all three zones (dorsal, intermediate (P10), though there was a trend toward a significant and ventral) of the rNTS. Local proliferation is the increase in the intermediate zone at P5 and the dorsal primary source of the microglia increase following CTX zone at P10. Taken together, these data may be in adult mice, which is supplemented by limited reflective of an immature immune response to neonatal migration (Bartel, 2012). When CTX is performed in adult CTX. The time course of the microglia response to mice, microglia numbers increase to about 200% of base- neonatal CTX is not yet established, which allows the line levels three days post-surgery and 250% of initial val- possibility that the lack of a significant microglia density ues five days after denervation in the rostral–central NTS increase across all rNTS zones could be explained by (Bartel, 2012; Bartel and Finger, 2013). We found a differential timing in the response to injury. somewhat smaller increase of approximately 190% base- Developmental variations in microglia number following line levels in the rNTS four days after CTX. This difference peripheral nerve injury are consistent with previous may be attributed to the area of NTS examined: the rostral research (Graeber et al., 1998; Melzer et al., 2001). 1/3rd in the present study verses the rostral–central sub- Injury during development can increase microglial nucleus. While the rNTS includes areas where CT termi- expression of the phagocytic marker CD68 in other nations vary in density, the rostral–central subnucleus brain regions accompanying neurodegeneration contains the densest concentration of CT fibers. The (Goranova et al., 2015). Microglial CD68 upregulation is microglia increase juxtaposed alongside CT fibers would also observed accompanying neuronal loss following therefore be most concentrated in a smaller area which neonatal facial nerve axotomy (Graeber et al., 1998). may explain the difference in the microglia increase When the facial nerve is severed in adult rats, there is between the two studies. Additionally, baseline density an increase in CD68-negative microglia, and neurons of microglia appears to be higher in adult mice compared are preserved (Graeber et al., 1998). The CTX paradigm with rats (approximately 1.8 in mice (Bartel, 2012; Bartel exhibits similar developmentally dependent denervation, and Finger, 2013) versus 1 in adult rats per with neonates exhibiting profound, permanent loss of 200,000 lm3 in the present study), indicating species fungiform taste buds and adults recovering completely differences. after several weeks (St. John et al., 1995; Kopka et al., In adult rodents, microglia can be protective during 2000; Sollars et al., 2002; Sollars, 2005). Surprisingly, chronic injury (Rice et al., 2015) and have been linked no CD68+ cells were detected in the rNTS of surgical to regeneration (Shokouhi et al., 2010). This may explain animals, independent of surgical age. There are several why the robust microglia response to adult CTX across all possible explanations for this finding. The developmen- rNTS zones, such as noted here, is accompanied by a tally differential microglial CD68 expression observed fol- preservation of the CT terminal fields despite temporary lowing facial nerve axotomy occurred in the facial motor loss of afferent input (Reddaway et al., 2012). Microglia nucleus, adjacent to the associated somas (Graeber have been associated with peripheral nerve regeneration et al., 1998). In contrast, the current study examined the in other systems (DeFrancesco-Lisowitz et al., 2015) and transganglionic microglia response. It is possible that the protective effect of microglia following adult CTX is CD68+ cells (non-microglia) may be expressed 4-days implicated by studies which impact microglia function. post CTX near the CT cell bodies in the geniculate gan- Minocycline has been used to inhibit microglial inflamma- glion, which may accompany cell death. tory response while leaving anti-inflammatory and protec- Graeber et al. (1998) observed CD68 expression con- tive functions intact (Kobayashi et al., 2013). As currently with soma loss. Following adult CTX in ham- minocycline has no impact on the microglia response after sters, degeneration of CT fibers in the rNTS occurs as adult mouse CTX (Bartel and Finger, 2013), this suggests early as 2-days post-surgery, but peaks 11 days-post that the microglia responding to adult CTX are already in (Whitehead et al., 1995). Therefore, it is possible that an anti-inflammatory and protective state, preserving the microglia upregulate CD68 at later timepoint than was CT terminal field (Reddaway et al., 2012) and possibly currently examined. This is supported by the observed facilitating axonal regeneration and the recovery microglia morphology more closely resembling a reactive, observed (St. John et al., 1995). transitionary morphology, as opposed to the more ame- boid morphology of fully activated, phagocytic, microglia. Future research is needed to determine if the expression CONCLUSIONS of CD68 in the rNTS following CTX may be delayed. Dif- The current study yields novel insight regarding microglia fering quantities and functions of microglia responding to morphology and density in the rat rNTS at four time points adult versus neonatal CTX may offer insight regarding dif- in CNS maturation. Early in life, microglia in the rNTS fering degrees of peripheral morphological and behavioral appear reactive, and are more numerous than in the recovery observed (Spector and Grill, 1992; St. John juvenile and adult rNTS. By P29, microglia are et al., 1995; St. John et al., 1997; Kopka et al., 2000; predominantly ramified and density has decreased to Sollars et al., 2002; Sollars, 2005). levels similar to adults (P54). Microglia’s role in synapse 46 A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 elimination (Paolicelli et al., 2011; Schafer et al., 2012) mechanosensory processing in the mouse nucleus tractus and the corresponding time course of CT terminal pruning solitarius. Chem Senses 41:515–524. (Sollars et al., 2006; Mangold and Hill, 2008) render Corson SL, Hill DL (2011) Chorda tympani nerve terminal field maturation and maintenance is severely altered following microglia a likely contributor to gustatory terminal field changes to gustatory nerve input to the nucleus of the solitary development. CD68-negative microglia assume a reac- tract. J Neurosci 31:7591–7603. tive morphology and increase in density in the CT terminal Crain JM, Nikodemova M, Watters JJ (2013) Microglia express field following CTX, independent of surgical age. The rel- distinct M1 and M2 phenotypic markers in the postnatal and adult ative increase in microglia after CTX is greater in older CNS in male and female mice. J Neurosci Res 91:1143–1151. animals, at ages associated with better functional recov- Dalmau I, Finsen B, Zimmer J, Gonzalez B, Castellano B (1998) Development of microglia in the postnatal rat hippocampus. ery. These findings highlight microglia as a potential reg- Hippocampus 8:458–474. ulatory mechanism underlying developmental plasticity in DeFrancesco-Lisowitz A, Lindborg JA, Niemi JP, Zigmond RE (2015) the gustatory system. This information adds to the grow- The neuroimmunology of degeneration and regeneration in the ing literature on the presence of microglia throughout peripheral nervous system. Neurosci 302:174–203. the brain, the important role of microglia during neural Delgado R, Carlin A, Airaghi L, Demitri MT, Meda L, Galimberti D, development, and the neuroimmunological response of Baron P, Lipton JM, Catania A (1998) Melanocortin peptides the brain following peripheral nerve injury. inhibit production of proinflammatory cytokines and nitric oxide by activated microglia. J Leukoc Biol 63:740–745. Fiske BK, Brunjes PC (2000) Microglial activation in the developing GRANTS rat olfactory bulb. Neurosci 96:807–815. Ginhoux F, Greter M, Leboeuf M, Nandi S, See P, Gokhan S, Mehler This work was supported in part by National Institutes of MF, Conway SJ, Ng LG, Stanley ER, Samokhvalov IM, Merad M Health, National Institute on Deafness and Other (2010) Fate mapping analysis reveals that adult microglia derive Communication Disorders #R01 DC04846 (S.I.S.) and from primitive macrophages. Science 330:841–845. by the University of Nebraska at Omaha Office of Glenn JA, Ward SA, Stone CR, Booth PL, Thomas WE (1992) Research and Creative Activity. The funding sources Characterisation of ramified microglial cells: Detailed morphology, morphological plasticity and proliferative capability. J Anat had no involvement in any aspect of the study, other 180:109–118. than providing financial support. Goranova V, Sifringer M, Endesfelder S, Ikonomidou C (2015) Neuroinflammation after traumatic injury to the developing brain. DISCLOSURES Scripta Scientifica Medica 47:47–52. Graeber MB, Lo´pez-Redondo F, Ikoma E, Ishikawa M, Imai Y, No conflicts of interest, financial or otherwise, are Nakajima K, Kreutzberg GW, Kohsaka S (1998) The declared by the authors. microglia/macrophage response in the neonatal rat facial nucleus following axotomy. Brain Res 813:241–253. Guagliardo NA, Hill DL (2007) Fungiform taste bud degeneration in AUTHOR CONTRIBUTIONS C57BL/6J mice following chorda-lingual nerve transection. J Study concept and design: A.J.R. and S.I.S. Acquisition of Comp Neurol 504:206–216. Hanisch UK, Kettenmann H (2007) Microglia: active sensor and data: A.J.R. and S.I.S. Drafting of the manuscript: A.J.R. versatile effector cells in the normal and pathological brain. Nat and S.I.S. Statistical analysis: A.J.R. Obtained funding: A. Neurosci 10:1387–1394. J.R. and S.I.S. Administrative, technical and material Hendricks SJ, Brunjes PC, Hill DL (2004) Taste bud cell dynamics support: S.I.S. Study supervision: S.I.S. during normal and sodium-restricted development. J Comp Neurol 472:173–182. Acknowledgments—The authors thank L. J. Martin for editing Hoffmann MB, Dumoulin SO (2015) Congenital visual pathway and comments, and J. M. Omelian, K. L. Apa, and M. M. Marian abnormalities: A window onto cortical stability and plasticity. for comments on previous versions of the manuscript. We thank Trends Neurosci 38:55–65. Howard RF, Walker SM, Mota PM, Fitzgerald M (2005) The ontology H. R. Bergwell for assistance with tissue processing. of neuropathic pain: Postnatal onset of mechanical allodynia in rat REFERENCES spared nerve injury (SNI) and chronic constriction injury (CCI) models. Pain 115:382–389. Hubel DH, Wiesel TM (1970) The period of susceptibility to the Abercrombie M (1946) Estimation of nuclear population from physiological effects of unilateral eye closure in kittens. J Physiol microtome sections. Anat Rec 94:239–247. 206:419–436. Alliot F, Godin I, Pessac B (1999) Microglia derive from progenitors, Imai Y, Ibata I, Ito D, Ohsawa K, Kohsaka S (1996) A novel gene iba1 originating from the yolk sac, and which proliferate in the brain. in the major histocompatibility complex class III region encoding Dev Brain Res 117:145–152. an EF hand protein expressed in a monocytic lineage. Biochem Avignone E, Lepleux M, Angibaud J, Na¨ gerl UV (2015) Altered Biophys Res Commun 224:855–862. morphological dynamics of activated microglia after induction of Jacquin MF, Zahm DS, Henderson TA (1993) Structure-function status epilepticus. J Neuroinflammation 12(202):1–13. relationships in rat brainstem subnucleus interpolaris. X. Bartel DL (2012) Glial responses after chorda tympani nerve injury. J Mechanisms underlying enlarged spared whisker projections Comp Neurol 520:2712–2729. after infraorbital nerve injury at birth. J Neurosci 13:2946–2964. Bartel DL, Finger TE (2013) Reactive microglia after taste nerve Kato G, Inada H, Wake H, Akiyoshi R, Miyamoto A, Eto K, Ishikawa injury: comparison to nerve injury models of chronic pain. T, Moorhouse AJ, Strassman AM, Nabekura J (2016) Microglial F1000Res 2:65. contact prevents excess depolarization and rescues neurons from Benjamini Y, Hochberg Y (1995) Controlling the false discovery rate: excitotoxicity. eNeuro 3. http://dx.doi.org/10.1523/ENEURO. A practical and powerful approach to multiple testing. J R Stat Soc 0004-16.2016. Series B Stat Methodol 57:289–300. Kettenmann H, Hanisch UK, Noda M, Verkhratsky A (2011) Breza JM, Travers SP (2016) P2X2 Receptor terminal field Physiology of microglia. Physiol Rev 91:461–553. demarcates a ‘‘transition zone” for gustatory and A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48 47

Kobayashi K, Imagama S, Ohgomori T, Hirano K, Uchimura K, interaction of the trigeminal and gustatory systems. Dev Neurobiol Sakamoto K, Hirakawa A, Takeuchi H, Suzumura A, Ishiguro N, 76:626–641. Kadomatsu K (2013) Minocycline selectively inhibits M1 Paolicelli RC, Bolasco G, Pagani F, Maggi L, Scianni M, Panzanelli P, polarization of microglia. Cell Death and Dis 4. http://dx.doi.org/ Giustetto M, Ferreira TA, Guiducci E, Dumas L, Ragozzino D, 10.1038/cddis.2013.54. Gross CT (2011) Synaptic pruning by microglia is necessary for Kopka SL, Geran LC, Spector AC (2000) Functional status of the normal brain development. Science 333:1456–1458. regenerated chorda tympani nerve as assessed in a salt taste Perea-Martinez I, Nagai T, Chaudhari N (2013) Functional cell types discrimination task. Am J Physiol Regul Integr Comp Physiol 278: in taste buds have distinct longevities. PLoS ONE 8:e53399. R720–31. Pulsinelli WA, Waldman S, Rawlinson D, Plum F (1982) Moderate Krimm RF, Hill DL (1997) Early prenatal critical period for chorda hyperglycemia augments ischemic brain damage: A tympani nerve terminal field development. J Comp Neurol neuropathologic study in the rat. Neurol 32:1239–1246. 378:254–264. Reddaway RB, Davidow AW, Deal SL, Hill DL (2012) Impact of Leake PA, Hradek GT, Synder RL (1999) Chronic electrical chorda tympani nerve injury on cell survival, axon maintenance, stimulation by a cochlear implant promotes survival of spiral and morphology of the chorda tympani nerve terminal field in the ganglion neurons after neonatal deafness. J Comp Neurol nucleus of the solitary tract. J Comp Neurol 520:2395–2413. 412:543–562. Reeves TM, Phillips LL, Povlishock JT (2005) Myelinated and Li Y, Yang J, Huang Y, Ren D, Chi F (2015) Shrinkage of ipsilateral unmyelinated axons of the corpus callosum differ in vulnerability taste buds and hyperplasia of contralateral taste buds following and functional recovery following traumatic brain injury. Exp chorda tympani nerve transection. Neural Regener Res Neurol 196:126–137. 10:989–995. Rice RA, Spangenberg EE, Yamate-Morgan H, Lee RJ, Arora RPS, Mangold JE, Hill DL (2007) Extensive reorganization of primary Hernandez MX, Tenner AJ, West BL, Green KN (2015) afferent projections into the gustatory brainstem induced by Elimination of microglia improves functional outcomes following feeding a sodium-restricted diet during development: Less is extensive neuronal loss in the hippocampus. Neurobiol Dis more. J Neurosci 27:4650–4662. 35:9977–9989. Mangold JE, Hill DL (2008) Postnatal reorganization of primary Schafer DP, Lehrman EK, Kautzman AG, Koyama R, Mardinly AR, afferent terminal fields in the rat gustatory brainstem is Yamasaki R, Ransohoff RM, Greenberg ME, Barres BA, Stevens determined by prenatal dietary history. J Comp Neurol B (2012) Microglia sculpt postnatal neural circuits in an activity 509:594–607. and complement-dependent manner. Neuron 74:691–705. Manning SM, Talos DM, Zhou C, Selip DB, Park HK, Park CJ, Volpe Schmued LC (1990) A rapid, sensitive histochemical stain for myelin JJ, Jensen FE (2008) NMDA receptor blockade with memantine in frozen brain sections. J Histochem Cytochem 38:717–720. attenuates white matter injury in a rat model of periventricular Schwarz JM, Sholar PW, Bilbo SD (2012) Sex differences in leukomalacia. J Neurosci 28:6670–6678. microglial colonization of the developing rat brain. J Neurochem Martin LJ, Sollars SI (2015) Long-term alterations in peripheral taste 120:948–963. responses to NaCl in adult rats following neonatal chorda tympani Sheng M, Cummings J, Roldan LA, Jan YN, Jan LY (1994) Changing transection. Chem Senses 40:97–108. subunit composition of heteromeric NMDA receptors during Matsui F, Kakizawa H, Nishizuka M, Hirano K, Shuo T, Ida M, Tokita development of rat cortex. Nature 368:144–147. Y, Aono S, Keino H, Oohira A (2005) Changes in the amounts of Shokouhi BN, Wong BZY, Siddiqui S, Lieberman AR, Campbell G, chondroitin sulfate proteoglycans in rat brain after neonatal Tohyama K, Anderson PN (2010) Microglial responses around hypoxia-ischemia. J Neurosci Res 81:837–845. intrinsic CNS neurons are correlated with axonal regeneration. May OL, Hill DL (2006) Gustatory terminal field organization and BMC Neurosci 11:13. developmental plasticity in the nucleus of the solitary tract Sollars SI (2005) Chorda tympani nerve transection at different revealed through triple fluorescent labeling. J Comp Neurol developmental ages produces differential effects of taste bud 497:658–669. volume and papillae morphology in the rat. J Neurobiol Melzer P, Savchenko V, McKanna JA (2001) Microglia, astrocytes, 64:310–320. and macrophages react differentially to central and peripheral Sollars SI, Smith PC, Hill DL (2002) Time course of lesions in the developing and mature rat whisker-to barrel morphological alterations of fungiform papillae and taste pathway: A study using immunohistochemistry for lipocortin1, buds following chorda tympani transection in neonatal rats. phosphotyrosine, s100b, and mannose receptors. Exp Neurol J Neurobiol 51:223–235. 168:63–77. Sollars SI, Walker BR, Thaw AK, Hill DL (2006) Age-related decrease Milev P, Maurel P, Chiba A, Mevissen M, Popp S, Yamaguchi Y, of chorda tympani nerve terminal field in the nucleus of the solitary Margolis RK, Margolis RU (1998) Differential regulation of tract is prevented by dietary sodium restriction during expression of hyaluronan-binding proteoglycans in developing development. Neuroscience 137:1229–1236. brain: Aggrecan, versican, neurocan, and brevican. Biochem Spector AC, Grill HJ (1992) Salt taste discrimination after bilateral Biophys Res Commun 247:207–212. section of the chorda tympani or glossopharyngeal nerves. Am J Mitani A, Watanabe M, Kataoka K (1998) Functional change of Physiol Regul Integr Comp Physiol 263:R169–R176. NMDA receptors related to enhancement of susceptibility to St. John SJ, Markison S, Guagliardo NA, Hackenberg TD, Spector neurotoxicity in the developing pontine nucleus. J Neurosci AC (1997) Chorda tympani transection and selective desalivation 18:7941–7952. differentially disrupt two-lever salt discrimination performance in Monfils MH, Driscoll I, Kamitakahara H, Wilson B, Flynn C, Teskey rats. Behav Neurosci 111:450–459. GC, Kleim JA, Kolb B (2006) FGF-2-induced cell proliferation St. John SJ, Markison S, Spector AC (1995) Salt discriminability is stimulates anatomical, neurophysiological and functional recovery related to number of regenerated taste buds after chorda tympani from neonatal motor cortex injury. Eur J Neurosci 24:739–749. nerve transection in rats. A J Physiol 269:R141–R153. Nikodemova M, Kimyon RS, De I, Small AL, Collier LS, Watters JJ Stevens B, Allen NJ, Vazquez LE, Howell GR, Christopherson KS, (2015) Microglial numbers attain adult levels after undergoing a Nouri N, Micheva KD, Huberman AD, Mehalow AK, Stafford B, rapid decrease in cell number in the third postnatal week. J Sher A, Litke AM, Lambris JD, Smith SJ, John SWM, Barres BA Neuroimmunol 278:280–288. (2007) The classical complement cascade mediates CNS Olsson B, Ridell B, Jerna˚ s M, Wadenvik H (2011) Increased number synapse elimination. Cell 131:1164–1178. of B-cells in the red pulp of the spleen in ITP. Ann Hematol Suzuki H, Oku H, Horie T, Morishita S, Tonari M, Oku K, Okubo A, 91:271–277. Kida T, Mimura M, Fukumoto M, Kojima S, Takai S, Ikeda T Omelian JM, Berry MJ, Gomez AM, Apa KL, Sollars SI (2016) (2014) Changes in expression of aquaporin-4 and aquaporin-9 in Developmental time course of peripheral cross-modal sensory optic nerve after crushing in rats. PLoS ONE 9(12):e114694. 48 A. J. Riquier, S. I. Sollars / Neuroscience 355 (2017) 36–48

Takahashi M, Vattanajun A, Umeda T, Isa K, Isa T (2009) Large- Webster SD, Park M, Fonseca MI, Tenner AJ (2000) Structural and scale reorganization of corticofugal fibers after neonatal functional evidence for microglial expression of C1qRP, the C1q hemidecortication for functional restoration of forelimb receptor that enhances phagocytosis. J Leukoc Biol 67:109–116. movements. Eur J Neurosci 30:1878–1887. Whitehead MC, McGlathery ST, Manion BG (1995) Tranganglionic Towfighi J, Mauger D, Vannucci RC, Vannucci SJ (1997) Influence of degeneration degeneration in the gustatory system consequent to age on the cerebral lesions in an immature rat model of cerebral chorda tympani damage. Exp Neurol 132:239–250. hypoxia-ischemia: A light microscopic study. Brain Res Dev Brain Williams RW, Rakic P (1988) Three-dimensional counting: An Res 100:149–160. accurate and direct method to estimate numbers of cells in Voorhies TM, Rawlinson D, Vannucci RC (1986) Glucose and sectioned material. J Comp Neurol 278:344–352. perinatal hypoxic-ischemic brain damage in the rat. Neurol Yager JY, Wright S, Armstrong EA, Jahraus CM, Saucier DM (2006) 36:1115–1118. The influence of aging on recovery following ischemic brain Wake H, Moorhouse AJ, Jinno S, Kohsaka S, Nabekura J (2009) damage. Behav Brain Res 173:171–180. Resting microglia directly monitor the functional state of synapses Zabel MK, Zhao L, Zhang Y, Gonzalez SR, Ma W, Wang X, Fariss in vivo and determine the fate of ischemic terminals. J Neurosci RN, Wong WT (2016) Microglial phagocytosis and activation 29:3974–3980. underlying photoreceptor degeneration is regulated by CX3CL1- Wall PL, McCluskey LP (2008) Rapid changes in gustatory function CX3CR1 signaling in a mouse model of retinitis pigmentosa. Glia induced by contralateral nerve injury and sodium depletion. Chem 64:1479–1491. Senses 33:125–135. Zhang LI, Bao S, Merzenich MM (2002) Disruption of primary auditory Watson Jr RE, Wiegand SJ, Clough RW, Hoffman GE (1986) Use of cortex by synchronous auditory inputs during a critical period. cryoprotectant to maintain long-term peptide immunoreactivity Proc Natl Acad Sci USA 99:2309–2314. and tissue morphology. Peptides 7:155–159.

(Received 9 September 2016, Accepted 24 April 2017) (Available online 04 May 2017)