Histone deacetylase inhibitors preserve function in aging axons


Address correspondence and reprint requests to Selva Baltan, Department of Neurosciences, Lerner Research Institute, The Cleveland Clinic Foundation, 9500 Euclid Avenue, NC30, Cleveland, OH, 44195 USA. E-mail: baltans@ccf.org


Aging increases the vulnerability of aging white matter to ischemic injury. Histone deacetylase (HDAC) inhibitors preserve young adult white matter structure and function during ischemia by conserving ATP and reducing excitotoxicity. In isolated optic nerve from 12-month-old mice, deprived of oxygen and glucose, we show that pan- and Class I-specific HDAC inhibitors promote functional recovery of axons. This protection correlates with preservation of axonal mitochondria. The cellular expression of HDAC 3 in the central nervous system (CNS), and HDAC 2 in optic nerve considerably changed with age, expanding to more cytoplasmic domains from nuclear compartments, suggesting that changes in glial cell protein acetylation may confer protection to aging axons. Our results indicate that manipulation of HDAC activities in glial cells may have a universal potential for stroke therapy across age groups.

Abbreviations used

artificial CSF


compound action potential


cyan fluorescent protein


glial fibrillary acidic protein


histone deactylase


mouse optic nerve




oxygen–glucose deprivation


suberoylanilide hydroxyamic acid


subcortical white matter


white matter

The risk for stroke increases drastically with age, and the central nervous system (CNS) white matter (WM) becomes intrinsically more vulnerable to oxygen–glucose deprivation (OGD) in older animals (Baltan et al. 2008; Baltan 2009). Several age-related changes in WM contribute to the increased susceptibility of axons to OGD. An up-regulation of GLT-1, the main Na+-dependent glutamate transporter in adult WM, correlates with early and robust release of glutamate causing an enhanced excitotoxicity (Baltan et al. 2008). Recently, we showed that Class I histone deacetylases (HDACs) are abundantly expressed in WM axons and glial cells, and pan- and Class I-HDAC inhibitors preserve WM structure and function during ischemia by conserving ATP and reducing excitotoxicity in young adult animals (Baltan et al. 2011a). Therefore, in this study, we investigated the potential of pan-HDAC inhibitor SAHA (suberoylanilide hydroxamic acid) and the Class I-HDAC inhibitor MS-275 (N-(2-aminophenyl)-4-[N-(pyridine-3yl-methoxy-carbonyl) aminomethyl] benzamide) for alleviating ischemic injury to aging axon function using an in vitro WM preparation of the isolated mouse optic nerve (MON). The optic nerve offers several advantages to study the mechanisms of aging WM injury, including the capacity to quantitatively assess axon function using electrophysiology and WM cellular components, and cytoskeleton using immunohistochemistry and confocal imaging (Baltan et al. 2008, 2011b).

Because the expression pattern of the HDACs in the aging CNS is unknown, we characterized the cellular expression of HDACs 1, 2, and 3, and in the aging mouse brain in primary and secondary cortex, CA1 regions of hippocampus, cingulate region, subcortical white matter, and optic nerve. Expression of Class I HDACs showed a cell- and region-specific pattern. Aging altered the extent of cellular HDAC expression in a region- and cell-specific manner compared with young CNS (Baltan et al. 2011a). The pan-HDAC inhibitor, SAHA, remarkably improved functional recovery of aging axons while preserving axonal mitochondria. The Class I inhibitor MS-275 comparably promoted axon function recovery after OGD in aging MONs. The results suggest that, in addition to young WM, HDAC inhibition also have non-transcriptional actions in axons and the distant processes of glial cells in aging WM, and may significantly modulate the response to injury in a cell- and region-specific manner. A detailed understanding of the mechanisms of HDAC inhibition on aging WM will be of central importance to investigate the therapeutic potential of HDACs's in stroke therapy.

Experimental procedures


SAHA and MS-275 were obtained from Selleck (Ontario, CA, USA). The mito-Cyan fluorescent protein (CFP) mice (Misgeld et al. 2007) were initially purchased from Jackson Laboratories (Bar Harbor, ME, USA) and bred at the University of Washington or Lerner Research Institute at the Cleveland Clinic Foundation. The Institutional Animal Care and Use Committee at both Institutions approved of all experimental procedures.

Recording techniques for optic nerve function and oxygen–glucose deprivation (OGD)

Mouse optic nerves (MONs) were obtained from male Swiss Webster mice and mice expressing mitochondrial-targeted CFP on a C57BL/6 background (Thy-1 CFP mice, Misgeld et al. 2007). Optic nerves were gently freed from their dural sheaths, placed in a Haas-top chamber superfused with artificial CSF (ACSF) and continuously aerated by a humidified gas mixture of 95%O2/5% CO2. All experiments were performed at 37°C. Suction electrodes back-filled with ACSF were used for stimulation and recording the compound action potential (CAP). The recording electrode was connected to an Axoclamp 900A amplifier (Molecular Devices, Sunnyvale, CA, USA) and the signal was amplified 50 times, filtered at 10 kHz, and acquired at 20 kHz. Stimulus pulse (50-μs duration, generated by PClamp 10, Molecular Devices) strength was adjusted to evoke the maximum CAP possible, and then increased another 25% for supramaximal stimulation. The MONs were allowed to equilibrate for at least 15 min in the chamber in normal ACSF. During experiments, the supramaximal CAP was elicited every 30 s. The OGD was induced by switching to glucose-free ACSF (replaced with equimolar sucrose to maintain osmolarity) and a gas mixture containing 95% N2/5% CO2. The OGD was applied for 60 min, glucose-containing ACSF and O2 were restored, and CAPs were recorded for up to 5–6 h.


This was performed in perfusion-fixed (4% paraformaldehyde in phosphate-buffered saline) MONs or coronal brain slices. Cryoprotection was achieved in increasing sucrose concentration (10–30% sucrose). Ten- to 16-μm-thick sections from each MON were blocked and permeabilized in 5–10% normal goat/donkey (50% by volume) serum (Sigma, St. Louis, MO, USA) and 0.3% Triton X-100 (Sigma) for 60 min at 23–25°C. Fifty-micrometer-thick sections from each brain were blocked and permeabilized in 5–40% normal goat/donkey (50% by volume) serum (Sigma) and 1% Triton X-100 (Sigma) for 60 min at room temperature. All primary antibodies were prepared in their respective solutions. Antibodies to HDACs 1–3 (monoclonal or polyclonal) were obtained from Sigma-Aldrich (St. Louis, MO, USA), and Sytox from Invitrogen (Carlsbad, CA, USA). Primary antibodies were used at a dilution of HDAC 1–3 1 : 250 and Sytox 1 : 25 000 and antibody specificity and control experiments were performed as previously reported (Baltan et al. 2011a,b). Brain sections were incubated in primary antibodies at 4°C for 2 days with constant shaking, while MONs were incubated in primary antibodies at 4°C overnight without shaking. After a thorough wash in phosphate-buffered saline, the tissue was exposed to a secondary antibody, prepared in 2% normal goat serum for overnight incubation. Donkey anti-rabbit Cy5, anti-mouse Cy5, anti-mouse Cy3, and anti-chicken Cy3, and anti-guinea pig (Jackson ImmunoResearch, West Grove, PA, USA) were used at 1 : 100 for MONs and at 1 : 250 dilution for free-floating brain sections.

Confocal microscopy and pixel-intensity measurements

The expression of HDCA1–3 was imaged using an Olympus (Center Valley, PA, USA) FV1000 upright laser-scanning confocal microscope. Sections were scanned with an argon and He–Ne laser for Cy3/543, eCFP/458, for Cy5/633, and for Cy2/488 fluorescence. Two to three adjacent sections from each MON were imaged for a total of three areas of interest per MON. A total of 10–16 optical sections of 1-μm thickness at 1024 × 1024 pixel size were collected in the z-axis from a single microscopic field using 40x (UPlanFL, oil immersion; numerical aperture 1.30; Olympus) objective lens under fixed gain, laser power, pinhole, and PMT settings. Images of brain slices were constructed by taking a total of 50 optical sections of 1-μm thickness at 1024 × 1024 pixel size in the z-axis using a 40x objective lens. To compare pixel intensity and quantify immunohistochemical staining, all sections were processed concurrently. Images were acquired with Olympus FluoView imaging software in sequential mode using multiple channels simultaneously. Z-stacks were projected into a single plane image before analysis and assessment of pixel intensity or colocalization. CFP pixel intensity was analyzed in freshly sectioned tissue for all groups.

Data analysis

Optic nerve function was monitored quantitatively as the area under the supramaximal CAP. The CAP area is proportional to the total number of excited axons and represents a convenient and reliable means of monitoring optic nerve axon function (Cummins et al. 1979; Stys et al. 1991). Irreversible injury was measured by determining residual CAP area, normalized to control CAP area, 5–6 h after the conclusion of OGD. Data were normalized by setting the mean of initial baseline values (measured over 20 min) to a value of 1.0. Results from several nerves were pooled, averaged, and plotted against time. All data are presented as mean ± SEM. In time-course plots, data and SEM bars are only shown every 3 min for simplicity. The CFP pixel intensity measurements were normalized by dividing the experimental values by the average obtained from control experiments. Statistical significance was determined by one-way anova followed by Bonferroni's post hoc test (Sigma Stat 3.5, San Diego, CA, USA). The n values indicate number of optic nerves in all figures.


HDAC expression pattern is region- and cell-specific in the aging brain

In coronal brain slices, we evaluated the expression and cellular localization of HDACs 1, 2, 3 (Fig. 1) using immunohistochemistry with isoform-specific antibodies and confocal imaging. Images (taken by 40x objective—see Methods) revealed the expression of HDAC 1, 2, and 3 to be widespread and prominent in primary and secondary motor cortex, (Fig. 1, CX), cingular region (Fig. 1, Cg), subcortical white matter (Fig. 1, SCWM), and CA 1 hippocampal regions (Fig. 1, CA 1) of the aging brain. The regional and cellular expression of Class I HDACs in aging brain is summarized in Table 1 in comparison to Class I-HDAC expression in young brain.

Figure 1.

Histone deacetylase (HDACs) 1, 2, and 3 are expressed by neurons and glial cells in aging brain slices and optic nerve. (Brain Column) HDAC 1 (red) was extensively expressed in neuronal Sytox (+) nuclei (blue), their cytoplasm and axons (white arrows) in the cortex (CX) and CA 1 region of hippocampus (CA 1). In subcortical white matter (SCWM), HDAC 1 principally outlined astrocyte nuclei, cytoplasm, and their proximal dendrites (white arrow heads). Some other glial nuclei (asterisk) in SCWM also showed HDAC 1 immunoreactivity. HDAC 2 (red) was intensely expressed in gray matter (cortex, cingulate region, and CA 1 region of hippocampus) or SCWM astrocytes (white arrows) and in their end-feet. HDAC 2 was consistently observed in neuronal nuclei (blue). HDAC 3 (red) was primarily expressed in neurons, and primary dendrites (yellow arrows) and axons (white arrows) in cortex, SCWM and CA 1 region of hippocampus. HDAC 3 labeling traced astrocyte processes in SCWM (white arrow heads) and end-feet outlining the vasculature (yellow arrow heads) in hippocampus. Scale bar = 200 μm. (Optic Nerve Column) HDAC 1 (red) labeling was prominent in aging axons (white arrows) and some Sytox (+) astrocyte nuclei (white asteriks) and processes. HDAC 2 (red) expression was most profuse in the aging optic nerve and reliably outlined axons throughout their length, weaving in and out of the thickness of optic nerve (white arrows). Asteriks point to glial nuclei (red + blue = purple) that show characteristic pattern of beads on a string for oligodendrocytes. HDAC 3 (red) immunoreactivity was mostly punctate in nature along axons (white arrows) and colocalized with some glial nuclei (blue, white asterisk) (See Table 1 for details). Scale bar = 50 μm.

Table 1. Comparing cell and region specific expression of Class I HDAC isoforms in the young and old CNS
 Gray MatterWhite Matter
  1. +, expressed

  2. −, not expressed

  3. ++, age-dependent change

  4. −−, age-dependent change

Astrocyte+ +  −−++
Nuclei+ +   ++
Cytoplasm+ +   ++
End-feet+ +   + 
End-feet++++  + 
End-feet+++++ ++++

The expression of HDAC 1 (Fig. 1, brain column, upper panel, red) colocalized with Sytox (+) neuronal nuclei (Fig. 1, brain column, upper panel, blue) completely outlining the cytoplasm of the primary and secondary motor cortical neurons. Immunoreactivity also consistently outlined the proximal axons. Consistent with this in the cingular region, HDAC 1 (+) axons extended from the motor cortex to SCWM (Fig. 1, brain column, upper panel, white arrows). In the hippocampus, nuclei and cytoplasm of CA1 pyramidal cells and interneurons expressed abundant levels of HDAC 1. In addition to neuronal expression, HDAC 1 was expressed in the nuclei and cytoplasm of cells that exhibited characteristic morphology of astrocytes in the subcortical white matter (SCWM) (Fig. 1, brain column, HDAC 1, white arrow heads). In triple-labeled sections, these cells proved to be glial fibrillary acidic protein (GFAP) (+) astrocytes (data not shown) . In the SCWM, HDAC 1 immunoreactivity was occasionally noted in cells (mostly nuclear, Fig. 1, asterisk) in addition to astrocytes and these were identified to as APC (+) oligodendrocytes (data not shown). HDAC 1 expression did not outline the dendrites in the hippocampus.

The expression of HDAC 2 (Fig. 1, brain column, middle panel, red) was principally nuclear in primary and secondary motor cortex, and CA1 pyramidal cell layer. The nuclear nature of HDAC 2 labeling was further supported by the lack of axonal labeling in the cingular region. Interestingly, HDAC 2 was abundant in nuclei, cytoplasm, and along entire processes of astrocytes in the hippocampus, SCWM, and even cingular region (Fig. 1, brain column, middle panel, white arrow heads).

The expression of HDAC 3 (Fig. 1, brain column, lower panel, red) was the most abundant in the aged brain with a mixed expression pattern. In neurons, HDAC 3 was colocalized with Sytox (+) nuclei, but also outlined the cytoplasm and the proximal axons. Consequently, HDAC 3 outlined axons of motor cortical neurons extending to SCWM, albeit to a greater extent than HDAC 1. It also labeled en-passant axons in SCWM (Fig. 1, brain column, lower panel, white arrows). Astrocytes expressed HDAC 3 in their nuclei and cytoplasm in the cingular region, SCWM, and hippocampus (Fig. 1 brain column, lower panel, arrow heads). HDAC 3 expression was more punctate in the processes of astrocytes in the hippocampus, but yet clearly outlined the end-feet surrounding the capillaries (Fig. 1 brain column, lower panel, yellow arrow heads). HDAC 3 labeled proximal dendrites of CA 1 hippocampal pyramidal neurons (Fig. 1 brain column, lower panel, yellow arrows).

HDAC 1, 2, and 3 are selectively expressed in axonal and glial cell compartments in aging optic nerve

We characterized the cellular expression pattern of HDACs 1, 2, and 3 in optic nerve from 12-month-old mice using immunohistochemistry and isoform-specific antibodies in conjunction with confocal imaging to support biological basis for HDAC inhibitor action in the aging axons (Fig. 1 Optic nerve column).

The expression of HDAC 1 labeling primarily outlined axons (Fig. 1, optic nerve column, upper panel, white arrows) and colocalized with some nuclei and occasionally with some processes of astrocytes (Fig. 1, optic nerve column, upper panel, asterisks). These observations were further confirmed in triple-labeled sections (Figure S1a, white arrows point to GFAP (+) astrocytes, yellow arrows show occasional HDAC 1 on astrocyte processes, Figure S1b, white arrows show HDAC 1 colocalize with NF-200 (+) axonal neurofilaments). The expression of HDAC 2 was abundant and principally restricted to axons with immunolabeling outlining entire length of axons weaving in and out of the thickness of the nerve (Fig. 1, optic nerve column, middle, white arrows). It also colocalized with some oligodendrocyte nuclei (Fig. 1, optic nerve column, middle, asteriks). The HDAC 3 labeling was lowest in aging optic nerves. Most of the labeling was observed in axons (Fig. 1, optic nerve column, middle, white arrows) and some glial nuclei (asterisk), but the distribution pattern was punctate in appearance. The extensive expression of these HDACs in glial cells, in addition to axons, implicated them as cellular targets of SAHA and MS-275 in aging WM.

HDAC inhibition preserves aging axon function during OGD

Because stroke incidence increases with age and CNS WM becomes intrinsically more vulnerable to ischemia in older animals (Baltan et al. 2008), the effects of HDAC inhibition on axon function and CFP pixel intensity after OGD were determined in MONs from 12-month-old Thy-1 mito CFP (Misgeld et al. 2007) or Swiss Webster mice (Figs 2 and 3). The functional integrity of optic nerve axons was monitored by quantifying the area under the compound action potentials (CAPs) evoked by supramaximal stimulus. After a 30-min baseline recording, MS-275 (N-(2-aminophenyl)-4-[N-(pyridine-3yl-methoxy-carbonyl)aminomethyl]benzamide), a specific Class 1-HDAC inhibitor, was introduced for 30 min and the superfusion conditions were maintained during 60-min OGD and the initial 30 min of reperfusion time. The axon function was then recorded for another 5–6 h after the end of HDAC inhibition. The onset of OGD caused a transient initial increase in the evoked CAP area before axon excitability rapidly diminished (Fig. 2). Following OGD, the CAP area recovered to 0 ± 0.5% (n = 5) of the maximum recorded CAP area. MS-275 (1 μM) prevented the complete loss of CAP area during OGD (22.3 ± 16.9%) and improved CAP area recovery to 69.5 ± 10.2% (n = 6, p < 0.001 compared to OGD, one-way anova).

Figure 2.

Class 1-Histone deacetylase (HDAC) inhibitor, MS-275 promote functional recovery of aging mouse optic nerve (MON). (a) MS-275 (1 μM) preserved some axon function during oxygen–glucose deprivation (OGD) (22.3 ± 16.9%) and (b) improved the CAP area recovery after OGD in aging MONs from Swiss Webster (SW) mice to 69.5 ± 10.2% (= 5) compared with OGD alone (0 ± 0.5%, n = 5). ***< 0.0001, one-way anova.

Figure 3.

Pan-Histone deacetylase (HDAC) inhibitor, suberoylanilide hydroxyamic acid (SAHA) improves axon function recovery in mouse optic nerves (MONS) from older mice. (a) SAHA (5 μM) pre-treatment preserved axon function (42.0 ± 10.1%, n = 6, open triangles) during oxygen–glucose deprivation (OGD) and (b) promoted CAP recovery (82.6 ± 2.4%, n = 6) to a similar extent as NBQX (85.6 ± 4.7%, n = 4, closed circles), an AMPA/kainate receptor blocker compared to OGD alone (6.3 ± 1.2%, n = 4, open squares) in Thy-1 mito CFP (+) mouse. ***p < 0.0001, one-way anova

HDAC inhibition preserves mitochondria in aging optic nerves

Because HDAC inhibition preserved axon function during ischemia by preserving mitochondria in optic nerves from young mice (Baltan et al. 2011a), we reasoned that protection of aging axon function during ischemia by HDAC inhibition should similarly preserve aging mitochondria. Axonal mitochondria in MONs can be selectively imaged in the mito-CFP mouse in which a neuron-specific Thy-1 promoter drives expression of mitochondrial-targeted CFP ((Misgeld et al. 2007), Fig. 4a). Therefore, we performed some experiments on 12-month-old Thy-1 mito CFP mice to fluorescently identify of mitochondrial shape, size, and location using SAHA (suberoylanilide hydroxamic acid), a Class I- and II-HDAC blocker. Axonal mitochondria in 12-month-old MONs appeared more abundant based on the increased levels of CFP fluorescence with longer and thicker mitochondria compared with young MONs (Fig. 4a, Control, inset). Interestingly, the reduction in CFP pixel intensity induced by OGD was substantially less (64.8 ± 2.5%, n = 4, p < 0.001, one-way anova) than that observed in 2-month-old mice [19.5 ± 0.5%, (Baltan et al. 2011a)], suggesting that the decline in CFP fluorescence observed in 2-month-old MON was not generally attributable to bleaching of the fluorophore in response to oxidative stress. In addition, OGD resulted in the formation of longer, thicker mitochondria in aging MONs contributing to enhanced CFP fluorescence after OGD (Fig. 4a, OGD, yellow arrows). As expected, the extent of irreversible injury detected after OGD was greater in MONs from 12-month-old Thy-1 mito mice (Fig. 3a, 6.3 ± 1.2%, n = 4), albeit slightly more than axon recovery observed from age-matched Swiss Webster mice (0 ± 0.5%, n = 5, Fig. 2) as previously reported (Baltan et al. 2011a). Conditions that ameliorated axon injury in older MONs, such as blockade of AMPA/kainate receptors with NBQX (Baltan et al. 2008) significantly preserved CFP signal (Fig. 4a, 80.0 ± 2.6%, n = 2, p < 0.05, one-way anova) and improved axon function recovery in older MONs (Fig. 3, 73.2 ± 1.8, n = 4, p < 0.001, one-way anova). Similarly, SAHA (5 μM) pre-treatment preserved CFP pixel intensity (Fig. 4a, 82.6 ± 2.4%, n = 6, p < 0.001, one-way anova) and axon function after OGD (Fig. 3, 85.6 ± 4.7%, n = 6, p < 0.001, one-way anova). Consistent with preservation of mitochondria, SAHA and NBQX prevented loss of the CAP area in the 12-month-old MONs during OGD. An identical and substantial proportion of axons remained functional during OGD with SAHA (42.0 ± 10.1%, n = 6) and NBQX [40.5 ± 1.9%, (n = 4), p = 0.90, Student's t-test] (Fig. 3a). The ischemic injury in older animals is predominantly mediated by glutamate-mediated excitotoxicity (Baltan et al. 2008). Similar amount of protection conferred by AMPA/kainate receptor blockade and HDAC inhibition on CFP pixel intensity and axon function in older animals suggests that HDAC inhibition blocks the excitotoxic pathway during WM ischemic injury.

Figure 4.

Suberoylanilide hydroxyamic acid (SAHA) preserves CFP (+) expression in mouse optic nerves (MONs) from older mice. (a) CFP (+) mitochondria were longer and thicker in MONs from 12-month-old Thy-1 CFP mice (inset; scale bar = 2 μm). Oxygen–glucose deprivation (OGD) consistently reduced CFP pixel intensity in 12-month-old MONs despite longer and brighter CFP (+) mitochondria (yellow arrows). (b) Blockade of AMPA/kainate receptors with NBQX (30 μM), or pan-Histone deacetylase (HDAC) inhibition with SAHA (5 μM), effectively preserved CFP pixel intensity during OGD. The preservation of CFP (+) mitochondria correlates with protection of the CAP area during OGD and profound recovery. *p < 0.05, ***p < 0.0001, one-way anova. Scale bar = 10 μm.


We directly tested and confirmed that HDAC inhibition was effective in protecting aging WM function against ischemic injury. This effect was apparent in two different strains of mice, suggesting that this is a general response of aging axons to HDAC inhibition. The altered expression pattern and cellular localization of HDAC 1, 2, and 3 in neuronal elements and WM components with age was intriguing and suggests that despite an age-dependent modification of HDAC expression, HDAC inhibition provides a universal protection of axon function against an ischemic episode in young (Baltan et al. 2011a) and old WM.

It is curious that aging caused most significant modification in expression of HDAC 3 by expanding to cellular compartments (note red markings, Table 1) such that neuronal cytoplasm and dendrites became HDAC (+) in addition to neuronal nuclei. Furthermore, astrocytes expressed HDAC 3 in their end-feet with aging. Because astrocyte end-feet are shown to contribute to vascular matrix degradation and repair following a stroke, these results may point to an important role for HDAC 3 in old brain. Note that HDAC 2 is the principle Class HDAC in young brain (Baltan et al. 2011a) expressed by the astrocyte end-feet (Table 1). While age-related changes in HDAC 1 and HDAC 2 expression were a mixture of increase and decrease, Class I-HDAC expression in SCWM remain unchanged with age (Table 1).

In contrast to the aging brain, HDAC 2 expression in aging MON showed the most remarkable change in its expression pattern. In the young MON, HDAC 2 faithfully labeled astrocytes nuclei, cell body, all processes, and end-feet (Baltan et al. 2011b), whereas in the old MON, HDAC 2 principally outlined axons and some glial nuclei which exhibited the beads-on-a-string pattern, a characteristic of oligodendrocytes nuclei. As a result, aging axons expressed both HDAC 1 and HDAC 2 as opposed to young axons solely expressing HDAC 1 (Baltan et al. 2011b). The nuclear export of HDAC1 in primary neuronal cultures is shown to mediate axonal beading and impede mitochondrial transport during glutamate exposure (Kim et al. 2010). Exposure to OGD leads to a robust up-regulation and translocation of HDAC 1-3 from nuclei to cytoplasmic domains of glial cells and axons in young MONs (Baltan et al. 2011a). Therefore, it is plausible that age-dependent translocation of HDAC 3 in the CNS and HDAC 2 in MON may play a role in increased vulnerability of tissue to ischemia. Although these findings point to change in cellular targets with age, Class I-HDAC inhibition conferred a profound protection to aging axons, proposing a universal protective role for HDACs for different age groups.

The impressive protection conferred by SAHA and MS-275 pre-treatment on aging axons propose that HDAC inhibition may interfere with glutamate release during OGD, similar to our recent report on young WM (Baltan et al. 2011b). Several lines of evidence support this proposal. First, vulnerability of aging WM to ischemia is primarily mediated by early and robust release of glutamate because of up-regulation of GLT-1 levels (Baltan et al. 2008). Ironically, GLT-1 up-regulation attenuates extracellular glutamate accumulation caused by OGD (Romera et al. 2004; Kawahara et al. 2005; Kosugi et al. 2005; Zhang et al. 2007) The cause of glutamate release during OGD is a loss of the transmembrane Na+ gradient, subsequently reversing Na+-dependent glutamate transport GLT-1 because of ATP depletion and failure of Na+–K+ ATPase. We showed that HDAC inhibition up-regulates GLT-1 levels and attenuate glutamate release by conserving ATP and subsequently reducing excitotoxicity in young MONs (Baltan et al. 2011a). It is therefore plausible that by preserving aging mitochondria, GLT-1 function may be kept in forward direction to effectively ameliorate excitotoxicity. Pan-HDAC inhibitors, such as SAHA, have been shown to enhance glutamate transport, increase expression of GLT-1 and GLAST in astrocytes (Wu et al. 2008; Allritz et al. 2009) and also GLT-1 expression in the striatum (Morland et al. 2004).

Further evidence in support of HDAC inhibition reversing the excitotoxic pathway in WM injury comes from the observation that a substantial portion of axons remained functional during ischemia when young and old MONs were pre-treated with SAHA. A similar protection of axon function during OGD is only observed when AMPA/kainate glutamate receptors are blocked with NBQX in young (Tekkok et al. 2007) and old WM (Baltan et al. 2008), further supporting our view that excitotoxicity is a major target of HDAC inhibition. We also confirmed our previous finding that WM is more vulnerable to OGD in older mice compared with young animals (Baltan et al. 2008) in this Thy-1 mitoCFP strain. The CAP area recovery only reached approximately 6% (in the absence of HDAC inhibitor), whereas young MON retained 13% of the CAP area. Because ischemic injury in older WM is mediated almost exclusively by excitotoxicity (Baltan et al. 2008), the beneficial effects of HDAC inhibition on aging axon function provides further confirmation that SAHA mitigates excitotoxic injury, although we cannot exclude additional mechanisms.

Another significant finding in our study was the demonstration that the axon-protective action associated with SAHA correlated with preservation of CFP (+) mitochondria in aging MONs. Although the extent of CFP (+) mitochondrial pixel intensity loss because of OGD was less pronounced in older WM, this was mainly because of the appearance of greatly enlarged mitochondria with brighter fluorescence. In keeping with this, HDAC inhibition and AMPA/kainate receptor blockade improved axon function recovery and restored CFP pixel intensity.

This study provides the first evidence that inhibition of Class I HDACs confers long-lasting benefits to CNS WM function in old animals following acute ischemic injury. Stroke occurs in the elderly and WM is injured in most strokes contributing to the disability associated with clinical deficits. By addressing HDAC inhibitor action in aged animals and specifically in WM, this study addresses one of the fundamental challenges in experimental stroke research. Based on our results, HDACs emerge as desirable therapeutic targets for stroke therapy in young (see (Gibson and Murphy 2010) and old population. These seminal findings may have direct clinical implications for treating various forms of brain injury.

Conflicts of interest

This work was supported by National Institute of Aging Grant AG033720 and the Lerner Research Institute. Richard S. Morrison and Sean P. Murphy are thanked for their various contributions. The author declares no conflict of interest.