Neuroprotein Targets of γ-Diketone Metabolites of Aliphatic and Aromatic Solvents That Induce Central

Abstract

Peripheral neuropathy associated with chronic occupational and deliberate overexposure to neurotoxic organic solvents results from axonal degeneration in the central and peripheral nervous system. Human and experimental studies show that axonopathy is triggered by the action of neuroprotein-reactive γ-diketone metabolites formed from exposure to certain aliphatic solvents (n-hexane, 2-hexanone) and aromatic compounds (1,2-diethylbenzene, 1,2-4-triethylbenzene, 6-acetyl-1,1,4,4-tetramethyl-7-ethyl-1,2,3,4-tetralin). Neuroprotein susceptibility is related primarily to their differential content of lysine, the ∊-amino group of which is targeted by γ-diketones. Specific neuroprotein targets have been identified, and the sequence of molecular mechanisms leading to axonal pathology has been illuminated. While occupational n-hexane neuropathy continues to be reported, lessons learned from its experimental study may have relevance to other causes of peripheral neuropathy, including those associated with aging and diabetes mellitus.

Keywords

1,2-diacetylbenzene;neuropathy axonopathy lysine neuroproteins diabetes mellitus

Introduction

Repeated workplace overexposure to n-hexane, an inexpensive organic solvent that has been widely used in industry, is an established cause of a self-limiting distal symmetrical peripheral neuropathy underpinned neuropathologically by central–peripheral distal axonal degeneration.^1,2 Other neurological phenomena associated with workplace exposure to n-hexane and solvent mixtures containing n-hexane include reduced balance control,³ impaired temperature sensation (cold allodynia),⁴ changes in color vision,^5,6 maculopathy and optic neuropathy,^7,8 and abnormal or reduced sense of smell.⁹ High-dose n-hexane overexposure, which alone or in a solvent mixture that has been used to induce euphoria, can precipitate a subacute onset of neuropathy resulting in severe muscle weakness and wasting.^10

-15

Peripheral neuropathy arising from repeated exposure to n-hexane was demonstrated 50 years ago in printing plants, sandal and shoe-making shops, and furniture factories in Asia, Europe, and the United States.^{10,16

-22} The problem has continued into the 21st century, with reports of n-hexane neuropathy among American automotive technicians,²³ shoe factory workers in Turkey,²⁴ screen printers in India,^25,26 electronic industry workers in Japan and South Korea,^27,28 and in a Chinese medicine plant in Taiwan.^29

-32 Additionally, between 1996 and 2004, there were 16 published reports of occupational n-hexane neuropathy in mainland China (Fujian, Guangdong, Heilongjiang, Henan, Jiangsu, and Liaoning) and, through 2009, 137 employees in a Suzhou electronic factory received hospital treatment for n-hexane neuropathy.³³ Sixty patients were treated for n-hexane intoxication in Guangdong Province between January 2017 and January 2018.³⁴ The main causes of chronic n-hexane poisoning (females > males) in an electronic enterprise of Guangdong’s Shenzhen City in 2017 were: long working hours, poor ventilation of workshops, unsatisfactory personal protective equipment, inadequate occupational health management, and an imperfect occupational health examination mechanism.³⁵

n-Hexane is metabolized by cytochrome P4502E1 (CYP2E1) to compounds with progressively greater neurotoxic potency, including 2-hexanol, 2-hexanone, and 2,5-hexanedione,^36
-38 a 1,4-diketone (γ-diketone) that binds directly with the ∊-amino group of lysine to form a 2,5-dimethyl pyrrole adduct (Figure 1).^40
-42 Pyrrole formation is an absolute requirement for neurotoxicity resulting in peripheral neuropathy.^43
-45 Similarly, 1,2-diethylbenzene, a solvent used in ion exchange resins and in powderless etching, appears to be metabolized to 1,2-diacetylbenzene (1,2-DAB),^46,47 which reacts with lysine to form a benzo-fused pyrrole or isoindole. The isoindole and pyrrole adducts form oxidized dimers and polymers that are seen as blue-violet and orange-colored pigments, respectively (Figure 1). Thus, rodents treated systemically with 1,2-DAB develop a bluish discoloration of skin, eyes, and internal organs, including nerves, spinal cord and brain, which reflects the reaction between the γ-diketone and proteins throughout the body. Calculations based on the density functional theory indicate the chromophore formed from the reaction of 1,2-DAB with proteins is likely to be composed of dimers of oxidized isoindoles.⁴⁸ A yellow discoloration results from systemic treatment of rodents with the aliphatic γ-diketone, 2,5-hexanedione (2,5-HD), which produces the lowest number of adducts in (rat) serum and the highest in the liver, where it has potential toxicity.⁴²

Figure 1.

γ-Diketone reactivity with lysine. Reaction of 2,5-hexanedione (2,5-HD, upper) and 1,2-diacetylbenzene (1,2-DAB, lower) with lysine to form dimethyl-pyrrole and -isoindole adducts, respectively, which form dimers that oxidize and polymerize to cross-link targeted proteins. The colored pigments that result appear to be associated with the dimer formation. Antioxidants block 1,2-DAB-induced cytotoxicity of SH-SY5Y cells.³⁹

Whereas 2,5-HD and 1,2-DAB are protein-reactive, chromogenic, and induce central–peripheral axonopathy, closely related isomers of the aromatic γ-diketone (eg, 1,3-DAB) and the aliphatic γ-diketone (eg, 2,4-HD) do not react with proteins to form chromogens and do not affect nerve fibers in the central nervous system (CNS) and peripheral nervous system (PNS).^49

-53 It can therefore be inferred that protein (lysine) reactivity is directly related to γ-diketone neurotoxic potential, and chromogenicity is a biomarker of neurotoxic action. Unknown is whether the same molecular mechanism underlies the ability of 2,5-HD to inhibit proliferation of neuroprogenitor cells in primary neuronal cultures, disrupt murine hippocampal neurogenesis,⁵⁴ and impair neurodevelopment of chick embryos.⁵⁵

The neurotoxic potency of γ-diketones increases stepwise from 2,5-HD, 3-methyl-2,5-HD, 3,4-dimethyl-2,5-HD (DMHD) and 1,2-DAB (Figure 1), presumably because the molecular configuration stepwise increases access of the reactive 1,4-diketo group to the target ∊-amino groups of protein lysines.^50,56 Similarly, 1,2-diethylbenzene (but not 1,3-DEB) and 1,2,4-triethylbenzene (but not 1,3,5-TEB), the parent compounds of 1,2-DAB and 1,2,4-DAB, respectively, induce peripheral neuropathy in rodents.⁵³ Other aromatic compounds, such as the polycyclic musk 6-acetyl-1,1,4,4-tetramethyl-7-ethyl-1,2,3,4-tetralin (AETT), formerly used in fragrances and as a food additive, yield a metabolite that is protein-reactive, forms a bluish chromogenic pigment, and induces nerve fiber damage in systemically treated rodents.^57,58 By contrast, neither tetralin nor its α-tetralol metabolite form a chromogen in vitro or induce axonal degeneration in mice treated by repeated intraperitoneal injection with either compound.⁵⁹ With the exception of tetralin, other aromatic hydrocarbons reported in early studies to cause blue-green-colored urine have not been studied for chronic neurotoxic potential (Table 1).

Table 1.

Chromogenic Organic Solvents as Reported by Gerarde⁶⁰ From His Experimental Animal Studies Using a Single Subcutaneous Injection of the Listed Substances of Unknown Purity in a Dose of Approximately 5 mL/kg.^a

Monocyclic	Dicyclic
1,2-Xylene	Indene
1,2-Ethyltoluene	Tetralin
1,2-Diethylbenzene	Diphenyl
1,3-Diethylbenzene	Diphenylmethane
1,2-Diisopropylbenzene	1-Methylnaphthalene
Triethylbenzene (mixture)	2-Methylnaphthalene
Diethyldisopropylbenzene	1-Ethylnaphthalene
Indane	2-Ethylnaphthalene

^a Gerarde reported that treatment of male and female rats and hamsters with 1,2-diethylbenzene (1,2-DEB) or 1,3-diethylbenzene (1,3-DEB), but not 1,4-diethylbenzene (1,4-DEB), caused the excretion of a blue dye in the urine. The sclera, blood, plasma, and tissues were stained a deep blue that persisted for many days. Later work using purified compounds found that 1,2-DEB (95%) and DEB mixtures,⁶¹ but not 1,3-DEB (99%) or 1,4-DEB (96%) alone, were chromogenic in rats given repeated oral doses of 500 and 750 mg/kg/d, 5 d/wk for 10 weeks.^47,48 Humans may also excrete colored (green) urine following exposure to tetralin,⁶¹ and a blue discoloration of the skin and internal organs develops in rodents repeatedly treated with acetylethyltetramethyltetralin.^57,58

The neuropathologic lesion induced by repeated systemic treatment with γ-diketones is one in which large-diameter and elongate nerve fibers in the CNS (mostly ascending and descending spinal cord tracts), as well as the PNS, initially develop axonal changes.⁶² Small-diameter and unmyelinated fibers are less susceptible but also exhibit degenerative changes. Experimental studies with rodents show that n-hexane, 2-hexanone, and 2,5-HD each induces degenerative changes in the distal regions of limb nerves and ascending and descending spinal cord tracts,^21,62 with comparable changes in the optic nuclei and hypothalamus.^63,64 Characteristic nerve fiber changes consist of focal axonal swellings filled with 10 nm neurofilaments (NFs) that commonly develop on the proximal sides of nodes of Ranvier, thereby resulting in retraction of paranodal myelin followed by localized demyelination and remyelination, with distal, retrograde (“dying-back”) axonal degeneration.²¹

Neurofilaments are synthesized in nerve cell bodies and transported distally (anterograde) along the axon at a rate of approximately 1 mm/d during which they are continuously exposed to a blood-borne neurotoxic agent. The spatiotemporal evolution of these pathological changes has been observed directly in organotypic cultures composed of structurally and functionally coupled explants of spinal cord, spinal ganglia, and striated muscle during continuous treatment (2.8 mM) with n-hexane or metabolites (2-hexanol, 2,5-hexanediol, 2-hexanone, 5-hydroxy-2-hexanone, and 2,5-HD) and 2,4-HD as a negative control.⁶⁵ Comparable treatment with the most potent lysine-reacting aliphatic γ-diketone, 3,4-dimethyl-2,5-hexanedione (DMHD),⁶⁶ induced giant axonal swellings filled with NFs in more proximal regions of peripheral nerve fibers in culture, followed later by distal, retrograde axonal degeneration.⁶⁵

The neuropathological picture is modified by the rate of systemic intoxication (dose, frequency, and duration) and the timing of examination. For example, axonal atrophy is prominent in peripheral nerves and ascending (gracile fasciculus and spinocerebellar tract) and descending (corticospinal and rubrospinal tracts) spinal tracts of rats treated by gavage with very high (up to 400 mg/kg/d) doses of 2,5-HD,^67,68 probably because the chemical assault rapidly arrests the anterograde transport of NFs, the density and subunit composition of which regulate axon diameter.⁶⁹ This is particularly evident when animals are treated with 1,2-DAB, a compound with markedly greater neurotoxic potency than 2,5-HD.^50,51 Animals treated with 1,2-DAB display massive proximal NF swellings, some of which develop intraspinally just distal to the initial axonal segment of the motor neuron.⁵⁰ When proximal axonal accumulation occurs in spinal roots, marked localized demyelination and remyelination results as a secondary response to axonal swelling and then attenuation, respectively. Indeed, the extent of demyelination and remyelination was so extensive in the spinal roots of AETT-treated animals that it was erroneously interpreted as a primary demyelinating disease.⁵²

The molecular basis of these compound-dependent patterns of γ-diketone-induced nerve fiber pathology appears to lie in their differential neurotoxic potency and, by extension, degree of protein adduction and polymerization. Highly potent neurotoxic γ-diketones, such as 1,2-DAB and DMHD, rapidly degrade 1 or more of the 3 proteins (NF heavy [NF-H], NF medium [NF-M], and NF-light [NF-L]) that make up the formed NF and/or the transport system on which anterograde NF transport depends. By contrast, lower potency γ-diketones, such as 2,5-HD, appear to complex with and degrade axonal proteins more slowly, such that the NFs are able to move anterograde along the axon until further transport is impeded in distal regions of the nerve fiber where they accumulate proximal to nodes of Ranvier. The rapid segregation of axonal microtubules from NFs induced by intrafascicular injection of 2,5-HD into the sciatic nerve in vivo may perturb the transport of NFs,⁷⁰ which then become disorganized and later accumulate in focal axonal swellings. The proposed relationship between the chemical structure of γ-diketones, relative potency, and the resulting location of focal axonal swellings with amassed 10 nm NFs is shown in Figure 2.

Figure 2.

Nerve fiber response to γ-diketones. Diagram of the location of initial focal axon (orange) swellings and associated secondary myelin (black) retraction in an elongate, large-diameter peripheral nerve fiber originating from the neuron soma (left) and terminating at sites of muscle (green) innervation. γ-Diketone potency and associated neuroprotein reactivity increase from 2,5-hexanedione (2,5-HD, upper) to 1,2-diacetylbenzene(1,2-DAB, lower), where neurofilaments accumulate both in the neuronal soma, distal to the initial segment, and in proximal spinal roots (as shown). The differential spatial “rigidity” (ability to rotate in space) of the 1,4-diketo groups may dictate the protein reactivity and, hence, neurotoxic potency of the illustrated γ-diketones. This diagram illustrates the principle that the toxic potential of the γ-diketone dictates how far neurofilaments are transported from the neuron cell body along the nerve fiber before they accumulate in focal axonal swellings; the greatest transport distance representing the least potent (2,5-HD) and, hence, the slowest developing effect. The most potent γ-diketone, 1,2-DAB, rapidly reacts with axonal proteins, such that anterograde transport of neurofilaments is arrested before or shortly after their exit from the neuron cell body. This figure is constructed from the author’s examination of single teased nerve fibers, and light and electron micrographs of cross-sectioned nerves, from peripheral nerve fibers undergoing γ-diketone axonopathy.^{2,52,53,62,63,65,71
-73}

Assessment in vitro of the chromogenic reaction of amino acids with DAB isomers demonstrates that 1,3-DAB fails to react with lysine, ornithine, glycine, γ-aminobutyric acid, or l-cysteine. By contrast, with the notable exception of cysteine, 1,2-DAB reacts rapidly and progressively with these compounds, most notably with l-lysine (Table 2). Since γ-diketones are particularly reactive in vitro with the exposed ∊-amino group of lysine, the total lysine content of axonal proteins may be important in regard to their susceptibility to γ-diketone-induced adduct formation and subsequent cross-linkage of proteins, including NF and microtubule proteins.^74
-76 With regard to 1,2-DAB, the total lysine content of axonal proteins seems to correlate with their susceptibility in vitro, those rich in lysine (NF-H, NF-M, kinesin, dynein, and tau) showing greater vulnerability relative to those with low lysine content (NF-L and β-tubulin)⁵⁰ (Table 3). Kinesin and dynein are motor proteins responsible for anterograde and retrograde axonal transport of cargoes, respectively, and the microtubule-associated protein tau appears to spatially regulate the balance of microtubule-dependent axonal transport in neurons.⁷⁷ Rat spinal cord slices incubated with 1,2-DAB showed a concentration-dependent decrease in motor and cytoskeletal proteins not seen with 1,3-DAB. While dynein and tau appeared similarly affected by 1,2-DAB, kinesin was most impacted by the neurotoxicant. Repeated systemic treatment of rats with 1,2-DAB, but not 1,3-DAB, affected motor and cytoskeletal proteins of sciatic nerves to a greater degree than those in the spinal cord. Motor proteins that drive axonal transport anterogradely (kinesin) and retrogradely (dynein), cytoskeletal protein NF-M, which is slowly transported in the anterograde direction, and microtubule-associated protein, tau, were differentially impacted by 1,2-DAB.⁷⁸

Table 2.

Chromogenic Reaction of Selected Amino Acids With Neurotoxic (1,2-DAB) and Non-Neurotoxic (1,3-DAB) Isomers of Diacetylbenzene.^a

Diketone	1,2-Diacetylbenzene (1,2-DAB)		1,3-DAB
Amino acid	30 minutes	60 minutes	60 minutes
l-Lysine	3.22 ± 0.02	3.51 ± 0.02	0.00
Ornithine	2.80 ± 0.01	3.56 ± 0.01	0.00
Glycine	2.21 ± 0.02	3.32 ± 0.04	0.00
GABA	1.93 ± 0.03	3.02 ± 0.02	0.00
l-Cysteine	0.00	0.02	0.00

^a Values represent mean absorbance ± standard deviation of the purple/blue color read at 540 nm on a scanning spectrophotometer. In triplicate, 0.5 mL aliquots of amino acid solution (0.1 M phosphate buffer, pH 7.4) were incubated with 0.5 mL of 1,2-DAB or 1,3-DAB (1 mM) at 37°C for 30 or 60 minutes. l-lysine, glycine, and γ-aminobutyric acid (GABA) and ornithine, a nonprotein amino acid and higher homologue of l-lysine, were the most reactive of 22 tested amino acids. l-Cysteine was among the least reactive of the amino acids tested. Modified from Kim et al.⁴⁹

Table 3.

Total Lysine Content of Axonal Proteins.^a

Axonal Protein	Percentage Lysine Content
Stathmin	15.5
Neurofilament heavy chain (NF-H)	13-17
Neurofilament medium chain (NF-M)	11-15
Neurofilament light chain (NF-L)	6-7
β-tubulin	3-4
Kinesin	8-12
Dynein	8
Tau	7-8

^a Protein reactivity of 1,2-DAB and formation of high-molecular-weight adducts correlated with lysine content. Both in vivo and in vitro, rat NF-H and NF-M formed polymers more readily than either NF-L or tubulin. Other 1,2-DAB protein targets involved in anterograde axonal transport included rat microtubule-associated tau, the rat motor proteins kinesin and dynein, and the rabbit glycolytic enzymes glyceraldehyde phosphate dehydrogenase (GAPDH: 8% lysine) and lactate dehydrogenase (LDH: 7-8% lysine). Stathmin, which contains a very high lysine content, was not tested. Data from Kim et al.⁴⁹

γ-Diketone-Induced Changes in Brain Gene Expression

Toxicogenomic studies have determined changes in brain gene expression following single intraperitoneal treatment of male C57BL/6 mice with 1,2-DAB (50 mg/kg) versus 1,3-DAB (50 mg/kg) or vehicle (2% acetone in saline). One-hour posttreatment, significant (ie, <1.5-fold, q < 0.1) 1,2-DAB-upregulated genes included transcription factors/regulator (Zic1, Fus, Junb, Klf4, Mapk8ip); those involved in cell cycle progression/proliferation (Rif1 homologue, Junb, Btg2, Mapk8ip); transmitter-related function (Gabt4, BQ746452 [similar to Gabbr1], Cplx2); protein degradation (Ube2 s, AK034214 [similar to Dirc2]); lipid metabolism (BB32327894 [similar to Gpam]), and microtubule-binding (Eml2, which negatively regulates microtubule polymerization, and AK038316, similar to Dock9, a member of a family of genes coding for proteins that regulate the actin cytoskeleton). At 6 hours posttreatment, 1,2-DAB upregulated genes included transcription factors/regulator (Dsip1, Mafk), those involved in cell cycle progression/proliferation (Pmaip1, Nfkbia, Igfbp1, Dsipl, Mafk, Cdkn1a [P21]), apoptosis regulating (Pmaip1, Angpt14, Dsip1, Cdkn1a (P21)), lipid metabolism (Tts-2.2 homologue), inhibitor of protein degradation (Ct1a2b), and neuronal antioxidant (Mt2). These findings demonstrate a complex and temporally evolving pattern of brain transcriptional perturbation in response to systemic diketone treatment.

There is sparse information on changes in gene expression associated with long-term treatment with n-hexane and 2,5-HD, its protein-binding neurotoxic metabolite. Neurofilament subunit gene expression was modestly reduced (approximately 20%) in the dorsal root ganglia of rats with 2,5-HD neuropathy.⁷⁹ Chronic treatment of rats with 2,5-HD reportedly increased cellular apoptosis in the spinal cord of rats by downregulating nerve growth factor expression and subsequently repressing the phosphoinositide 3-kinase/Aktsignaling pathway.⁸⁰ However, long-term systemic treatment of rats with 2,5-HD reportedly increased the percentage of proliferating neural stem and progenitor cells in the granule cell layer of the hippocampus and the subgranular zone.⁸¹ Neuropathological changes were found in the mammillary body and visual nuclei of cats treated with low levels of 2,5-HD for prolonged periods.⁶³

n-Hexane inhalation altered the methylation status of promoters of genes associated with ovarian cell apoptosis and steroid hormone biosynthesis,⁸² and treatment with 2,5-HD increased apoptosis in human ovarian granulosa cells.^83,84

γ-Diketone-Induced Changes in Neuroprotein Abundance

Early CNS Protein Changes

The brain proteome was examined in mice recovering from a single intraperitoneal injection of 1,2-DAB, 1,3-DAB, or DAB vehicle (Table 4). Marked 1,2-DAB-associated changes were noted in the relative abundance of brain proteins involved in microtubule interaction, other structural proteins, and cell energy metabolism. The greatest change was seen for actin α-2. Association of actin filaments with microtubules is important for various cellular processes, such as cell division, migration, vesicle and organelle transport, and axonal growth.⁸⁵ Actin α-2 has at least 10 lysine-acetylation sites, and lysine-acetylated actin can interact with cyclase-associated protein to form an inhibitor of inverted formin 2 (INF2). Significantly, dominant missense mutations in INF2 are linked to Charcot-Marie-Tooth disease, an inherited human axonal neuropathy.^86,87

Table 4.

DAB Neuroproteome Relative Abundance of Brain Proteins of Young Adult C57/BL6 Mice 168 Hours After a Single Intraperitoneal Injection of 1,2-DAB (50 mg/kg Body Weight), 1,3-DAB (50 mg/kg), or DAB Vehicle (2% Acetone in Saline).^a

Protein Abundance	Change in Abundance		Relative Abundance
Protein Name/Function	1,2-DAB/1,3-DAB	1,2-DAB/Vehicle	1,2-DAB	1,3-DAB	Vehicle
Microtubule-interacting
Stathmin	6.98	8.85	108.54	15.55	12.27
Oral–facial–digital syndrome 1	4.68	∞	46.46	9.96	0.00
Restin (clip-170, clip-50)	3.38	5.71	8.62	2.55	1.51
Microtubule-associated protein 2	5.47	3.79	3.94	0.72	1.04
Synuclein (α + β)	4.12	2.09	2.72	0.66	1.30
Other structural proteins
Neurofilament medium (NF-M)	4.09	7.37	2.21	0.54	0.30
Neurofilament light (NF-L)	1.64	1.24	0.46	0.28	0.37
Actin alpha 1	2.30	1.60	18.77	8.15	11.73
Actin alpha 2	14.85	50.77	15.74	1.06	0.31
Procollagen type V1 alpha 2	3.29	5.38	5.16	1.57	0.96
Peripherin	3.84	2.32	1.23	0.32	0.53
Cellular energy metabolism
Malate dehydrogenase, soluble	23.59	75.19	24.06	1.02	0.32
Triosephosphate isomerase	2.20	3.97	18.10	8.22	4.56
Enolase-2 (neuronal)	1.73	∞	14.92	8.64	0.00
NADH dehydrogenase Fe-S protein 3	0.31	0.45	10.93	35.83	24.13
Phosphoglycerate kinase 2	5.34	10.69	7.48	1.40	0.70
ATP synthase β-chain	3.17	5.07	0.76	0.24	0.15
Vesicular tracking
Synapsin 1	∞	2.65	11.46	0.00	4.33
Kinesin family member 20A	38.13	∞	6.10	0.16	0.00

Abbreviations: ATP, adenosine triphosphate; NADH, nicotinamide adenine dinucleotide.

^a Half brains were pooled from 2 animals from each group. Each sample was homogenized in 1 mL 100 mM ammonium bicarbonate, 5 mM dithiothreitol, pH 7.8, centrifuged at 80,000 RPM for 1 hour in a TL-100 rotor, the supernatant collected and assayed for protein at a 1:100 dilution; 100 μg of each sample was digested with 1:50 trypsin to protein (wt/wt), for 1 hour at 37°C (20% ACN). Peptide cleanup was performed on a SP-C18 column and eluted in 80% acetonitrile (CAN), 0.1% acetic acid, and 0.01% trifluoroacetic acid. Each sample was run in duplicate on a 9-Tesla Fourier-transform ion cyclotron resonance mass spectrometer. The mouse brain VOXEL PMT database was used for peak matching of proteins of differing abundance. A total of 1643 redundant proteins was identified, of which 548 proteins were modulated by 1,2-DAB, with 196 mapped to cell networks. Performed in collaboration with J. Pounds and J. Adkins, Pacific Northwest National Laboratory, US Department of Energy, Richland, Washington, DC.

It seems reasonable to posit that 1,2-DAB attacks actin’s lysine sites, thereby blocking the formation of INF2 and contributing to the genesis of γ-diketone axonopathy. There was also a markedly increased abundance in 1,2-DAB versus 1,3-DAB or vehicle of CLIP-1 (CAP-GLY domain-containing linker protein), a microtubule plus-end-associated protein that interacts with dynein and binds tightly to formins to accelerate actin filament elongation.^88,89 1,2-Diacetylbenzene versus 1,3-DAB also showed some increased abundance of microtubule-associated protein 2 and α-synuclein, which facilitates the formation of short, mobile transportable microtubules that play an important role in axonal transport.⁹⁰ These changes in protein expression should be confirmed with additional proteomic studies.

Brains of mice 1 week after a single 1,2-DAB treatment showed increased abundance of stathmin (Table 4) and proteins involved in stathmin regulation: cyclin-dependent kinase 5 (cdk5), RAS-related C3 botulinum substrate 1 (rac1), heat shock protein conjugate 70 (hsv70), and protein phosphatase 2a (pp2a). These brains also showed changes in stathmin-regulated genes, including cyclin-dependent kinase inhibitor 1a (Cdkn1a, P21), epidermal growth factor (Egf), and Krüppel-like factor 4 (Klf4). Stathmin is a ubiquitous, brain-enriched phosphoprotein with a very high lysine content that binds to tubulin and inhibits microtubule assembly, an activity regulated by phosphorylation.^91,92 Stathmin is of special interest in relation to γ-diketone axonopathy because aged stathmin knockout mice also develop axonal degeneration and secondary demyelination.⁹³ Additionally, incubation of mouse brain and testes homogenates with 1,2-DAB (1-10 mM) reduced the intensity of the native stathmin band in a concentration-dependent manner, with corresponding increases in high-molecular-weight protein adducts. By contrast, under similar conditions, 1,3-DAB had no effect on stathmin, and adducts were not observed. Stathmin is therefore of potential mechanistic relevance to γ-diketone axonopathy.⁹⁴

Later CNS Protein Changes

Neuroprotein studies have been performed on the lumbosacral spinal cord of rats with severe γ-diketone neuropathy. 1,2-DAB significantly altered the expression of protein disulfide isomerase, an enzyme involved in protein folding, and gelsolin, an actin-capping and -severing protein. Thirty-four proteins were markedly modified by 2,5-HD, of which NF-L, gelsolin, protein disulfide isomerase, glutathione S-transferase, nicotinamide adenine dinucleotide (reduced) dehydrogenase 1α, pyruvate kinase, and fatty acid synthase were also modified by 1,2-DAB. 2,5-Hexanedione induced a 1,2-DAB-like proteomic signature by altering the expression levels of proteins involved in maintaining the physical integrity of axons (reduced), in controlling redox and protein-folding mechanisms (reduced), and in supporting energy metabolism (increased).⁵¹ While the spinal cord proteome also suggested a reduction in α-II spectrin (Spna2), a key protein in the maintenance of axonal integrity, degradation of Spna2 by calpain- and/or caspase is reportedly not central to the pathogenesis of 1,2-DAB axonopathy.⁹⁵

Perturbation of energy metabolism resulting in reduced adenosine triphosphate (ATP) has been postulated to be linked to the etiology of central–peripheral axonopathy.⁹⁶ 2,5-Hexanedione reduced the rate of ATP synthesis in isolated brain mitochondria⁹⁷ and pyruvate-restored ATP deficits in cat nerves treated with 2,5-HD.⁹⁸ Consistent with this hypothesis, recovery from systemic treatment with 1,2-DAB was associated with a very marked increase in brain abundance of soluble malate dehydrogenase and, to a lesser extent, other proteins involved in energy metabolism, including glycolysis (triosephosphate isomerase, phosphoglycerate kinase 2, enolase 2) and the electron transport chain (nicotinamide adenine dinucleotide dehydrogenase Fe-S protein 3; ATP synthase β-chain; Table 4). Oxidative stress and resulting tissue damage have been attributed to γ-diketone toxicity of liver, kidney, brain, and SY5Y neuroblastoma cells.^39,99,100

Normal Aging and Diabetes Mellitus

Are neurotoxic γ-diketones physiological metabolites and might they participate in other types of peripheral neuropathy, including those associated with the aging process and diabetes mellitus? One study of 31 normal subjects with no known n-hexane exposure found low levels (<0.006 mg/L) of free 2,5-HD in urine.¹⁰¹ A second study reported that healthy subjects without occupational exposure to n-hexane had detectable levels of 2,5-HD in the blood (6-30 μg/L) and urine (0.17 and 0.98 mg/L), only a minimal part of which was considered to have derived from exposure to hydrocarbon-polluted air.¹⁰² However, a third study found that 1.3% of 1200 normal subjects with no known occupational exposure to n-hexane had blood levels of the neurotoxic alkane above the method detection limit.¹⁰³ A fourth investigation of urine samples from 123 healthy Italian subjects recorded a 2,5-HD reference value of 0.795 mg/L for men and 0.627 mg/L for women.¹⁰⁴ A fifth very large study of healthy Chinese subjects (n = 8235) with no occupational exposure to n-hexane or 2-hexanone showed a median urine 2,5-HD concentration of 0.171 mg/L for males and 0.147 mg/L for females, with increasing 2,5-HD excretion with the advance of age.¹⁰⁵ In a sixth study, investigation of 208 male and female subjects revealed a median level of urinary pyrrole adducts of unstated origin of 0.91 nmol/mL.¹⁰⁶ Finally, with respect to diabetes mellitus, a small study of serum samples from normal individuals, and from subjects with type 2 diabetic neuropathy, revealed similar qualitative profiles of volatile metabolites (low nanogram concentration), including relatively high and low concentrations, respectively, of 2-hexanone and 3-heptanone, both of which can undergo ω-oxidation to form γ-diketones.¹⁰⁷ Also present in similar concentrations in normal and diabetic sera was 2-butanone (methyl ethyl ketone), a compound that potentiates the neurotoxic potency of n-hexane and 2-hexanone.^108

-111 Since degenerative nerve fiber changes with the advance of aging resemble those seen in the early stages of distal axonopathies,¹¹² and central–peripheral axonopathy underpins type 2 diabetic neuropathy,⁷¹ it will be important to determine the origin of γ-diketones and precursors with neurotoxic potential in normal subjects and whether they contribute to nerve fiber changes in aged subjects and those with diabetic neuropathy.

Conclusion

Structure-activity studies^52,113 show that γ-diketone metabolites of certain aliphatic and aromatic solvents selectively react with proteins throughout the body to form chromogenic polymers. Despite this global assault and widespread formation of protein adducts (in serum, liver, kidney, and brain),¹¹⁴ other than the nervous system, few tissues (testes, liver, and kidney) undergo degenerative/atrophic changes in γ-diketone-treated animals.^99,115 The special susceptibility of CNS and PNS neurons to γ-diketones appears to be related to their requirement for transport of proteins over very long distances (ie, from cell soma via the axon to the nerve terminal). Although protected by blood–brain/nerve barriers, γ-diketones not only cross these regulatory interfaces but can also increase their permeability.¹¹⁶ Elongated, large-diameter nerve fibers are affected first, probably because of the large volume of NF and other proteins undergoing anterograde transport. However, increasing the γ-diketone potency and dosage results in more rapid protein adduction, such that NFs are arrested shortly after commencing their transport from sites of synthesis in neuronal somata. This results in the formation of focal giant axonal swellings containing masses of disoriented NFs, while more distal regions of axons undergo atrophy because they fail to receive a normal complement of NFs. Distal, retrograde axonal degeneration then ensues by a proposed mechanism in which trapped mitochondria release stored calcium and trigger calcium-activated proteases that effectively sever the axon distal to the site of NF accumulation.⁷²

The exact targets of γ-diketones are not clear. Microtubules are unequivocally impacted by γ-diketones, either directly or indirectly, or both. In vitro, 2,5-HD-treated tubulin had a decreased lag time and an increased maximal velocity of microtubule assembly.¹¹⁵ Systemic 2,5-HD treatment causes longitudinal aggregation of axonal microtubules, sometimes in isolated clusters and sometimes decorating the perimeter of mitochondria, an event that occurs within minutes of intraneural γ-diketone injection⁷⁰ and followed by the development of NF-laden axonal swellings days later.⁷³ This seems to be associated with loss of the 3- to 5-nm thick cross-bridges formed by MAP 2 or tau on axonal microtubules¹¹⁷ or from the carboxy-terminal tail domains of phosphorylated NF-H and NF-M that laterally interconnect intermediate NFs.^118
-120 Neurofilament medium (and NF-H?) contains a variable lysine–serine–proline (KSP) repeat subdomain flanked by 2 highly conserved subdomains, the number of tri-, tetra-, and pentapeptide KSP repeats dispersed along the subdomain increasing in proportion to axon length and caliber.¹²¹ Given the high affinity of γ-diketones for lysine residues, their reactivity with high lysine content NF-H and NF-M, the exceptional vulnerability of long and large-diameter axons in systemically treated rodents, and the earlier and greater involvement of lower relative to upper extremities in γ-diketone neuropathy, the NF sub-units NF-H and NF-M are also intimately involved in the pathogenesis of γ-diketone axonopathy. Additionally, NF-H knockout mice are refractory to the γ-diketone-like giant NF swellings induced by systemic treatment with β, β′-iminodipropionitrile, which suggests that NF-H is a key mediator of this type of axonopathy.¹²²

Normal anterograde axonal transport of NF subunits appears to be dependent on myosin-mediated interactions within the actin cortex that lines the inner axolemmal surface.¹²³ Actin is also involved in microtubule-dependent vimentin transport.¹²⁴ Actin contains 19 lysines of variable reactivity, of which K325 and K327 are most reactive with acetic anhydride.¹²⁵ Given the very large increased production of brain actin following systemic treatment with 1,2-DAB, actin lysines are also likely to be important γ-diketone targets. Stathmin, which binds to and sequesters monomeric tubulin, thereby preventing microtubule polymerization, has a very high lysine content and is thus a prominent target for γ-diketones, a position supported by the increase in brain abundance of stathmin after systemic treatment with 1,2-DAB. Serine 63, a key phosphorylation site for stathmin-induced destabilization of microtubules, is next to lysine 62,^92,126 which might be an important site for γ-diketone attack. Stathmin binds directly along the microtubule wall, such that a pool of this protein is immediately available to participate in protofilament dissociation when the moving plus end of a depolymerizing microtubule reaches stathmin molecules.¹²⁷ Other axonal proteins with significant lysine components are the motor proteins dynein and kinesin, although it is noteworthy that increased brain abundance of these neuroproteins was not apparent after systemic treatment with 1,2-DAB.

The foregoing description of γ-diketone neuropathy represents one of the most detailed analyses of the molecular mechanisms underlying a central–peripheral distal axonopathy, a pattern of nervous system damage seen in a number of genetic, metabolic, nutritional, and toxic neuropathies. The molecular mechanisms involved in the neurotoxic actions of γ-diketones may therefore have relevance in comparable neuropathologic states, such as those associated with aging and diabetes mellitus.

Footnotes

Acknowledgments

The experimental studies reviewed here were variously performed by students, technicians, postdoctoral fellows, and faculty colleagues at Albert Einstein College of Medicine, Oregon Health & Science University, Oregon State University, Pacific Northwest National Laboratory, and University of Houston; in surname alphabetical order: Joshua Adkins (proteomics), Monica Bischoff (animal studies), Alan Cranson (toxicogenomics), Max Deinzer (mass spectrometry), David Dixon (computation), Dickran Houropian (neuropathology), Seyed Hashemi (protein studies), Robert Kayton (electron microscopy), Min-Sun Kim (neuropathology), Michael Lasarev (biostatistics), Valerie Palmer (animal studies and toxicogenomics), the late Edith Peterson (tissue culture), Michael Politis (animal studies), Colin Poole (chromatography), Joel Pounds (proteomics), Mohammad Sabri (protein studies), Herbert Schaumburg (neurology and neuropathology), Mary Seelig (tissue culture), Dean Sproles (toxicogenomics), Arnold Sterman (neuropathology), Sarah Trimpkin (mass spectrometry), Desire Tshala-Katumbay (neuropathology and protein studies), Bellina Veronesi (tissue culture), Joy Zagoren (animal studies), Chang-Guo Zhan (computation studies), and the late Albert Zlatkis (chromatography). The author thank Chen Xiao for discussion. Ann Hubbs kindly reviewed the article.

Declaration of Conflicting Interests

The author declares no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author discloses receipt of the following financial support for the research, authorship, and/or publication of this article: The toxicogenomic and proteomic studies reported here for the first time were supported by grant U19 ES011384 from the National Institute of Environmental Health Sciences.

ORCID iD

Peter S. Spencer

References

Spencer

Schaumburg

Sabri

Veronesi

. The enlarging view of hexacarbon neurotoxicity. Crit Rev Toxicol. 1980;7(1):279–356.

Spencer

Schaumburg

. Ultrastructural studies of the dying-back process. IV. Differential vulnerability of PNS and CNS fibers in experimental central-peripheral distal axonopathies. J Neuropathol Exp Neurol. 1977;36(2):300–320.

Herpin

Gargouri

Gauchard

, et al. Effect of chronic and subchronic organic solvents exposure on balance control of workers in plant manufacturing adhesive materials. Neurotox Res. 2009;15(7):179–186. doi:10.1007/s12640-009-9018-0.

Pradhan

Tandon

. N-hexane neuropathy with vertigo and cold allodynia in a silk screen printer: a case study. Int J Occup Med Environ Health. 2015;28(1):915–919. doi:10.13075/ijomeh.1896.00327.

Chang

. Patients with n-hexane induced polyneuropathy: a clinical follow up. Br J Ind Med. 1990;47(4):485–489.

Beckman

Eisen

Bates

, et al. Acquired color vision defects and hexane exposure: a study of San Francisco bay area automotive mechanics. Am J Epidemiol. 2016;183(2):969–976. doi:10.1093/aje/kwv328.

Sobue

Yamamura

Ando

Iida

Takayanagi

. n-Hexane polyneuropathy outbreak among vinyl sandal manufacturers. Clin Neurol. 1968;8(8):393–403.

Chang

Fong

, et al. N-Hexane neuropathy in offset printers. J Neurol Neurosurg Psychiat. 1993;56(4):538–542.

Lee

Zhang

Chen

Lam

Wong

. Occupational exposure to mixtures of organic solvents increases the risk of neurological symptoms among printing workers in Hong Kong. J Occup Env Med. 2004;46(2):323–330.

10.

Goto

Matsumura

Inoue

Murai

Shida

. Toxic polyneuropathy due to glue sniffing. J Neurol Neurosurg Psychiat. 1974;37(1):848–853.

11.

Korobkin

Asbury

Sumner

Nielsen

. Glue-sniffing neuropathy. Arch Neurol. 1975;32(6):158–162.

12.

Altenkirch

Mager

. Toxic polyneuropathy after sniffing contact glue thinner [in German]. Dtsch Med Wochenschr. 1976;101(2):95–99.

13.

Smith

Alpers

. N-Hexane neuropathy due to rubber cement sniffing. Muscle Nerve. 1997;20(11):1445–1450.

14.

Becker

Lee

Troost

. Glue sniffing polyneuropathy: an underrecognized aspect of a public health hazard. J Adolesc Health. 2004;34(1):94–96.

15.

Huang

. Polyneuropathy induced by n-hexane intoxication in Taiwan. Acta Neurol Taiwan. 2008;17(1):3–10.

16.

Yamamura

. N-hexane polyneuropathy. Folia Psychiat Neurol Jpn. 1969;23(1):45–57.

17.

Herskowitz

Ishii

Schaumburg

. N-hexane neuropathy. A syndrome occurring as a result of industrial exposure. N Engl J Med. 1971;285(2):82–85.

18.

Abbritti

Siracusa

Cianchetti

, et al. Shoe-makers’ polyneuropathy in Italy: the aetiological problem. Br J Ind Med. 1976;33(2):92–99.

19.

Cianchetti

Abbritti

Perticoni

Siracusa

Curradi

. Toxic polyneuropathy of shoe-industry workers. A study of 122 cases. J Neurol Neurosurg Psychiat. 1976;39(6):1151–1161.

20.

Paulson

Waylonis

. Polyneuropathy due to n-hexane. Arch Intern Med. 1976;136(3):880–882.

21.

Buiatti

Cecchini

Ronchi

Dolara

Bulgarelli

. Relationship between clinical and electromyographic findings and exposure to solvents, in shoe and leather workers. Br J Ind Med. 1978;35(9):168–173.

22.

Huang

Chu

Shih

. Occupational neurotoxic diseases in Taiwan: a review of the outbreaks and clinical features. Changgeng Yi Xue Za Zhi. 1997;20(7):71–78.

23.

Harrison

Larabee

Con

, et al. N-Hexane–related peripheral neuropathy among automotive technicians—California, 1999-2000. Mort Morb Wkly Rep. 2001;50(2):1011–1013.

24.

Kutlu

Gomceli

Sonmez

Inan

. Peripheral neuropathy and visual evoked potential changes in workers exposed to n-hexane. J Clin Neurosci. 2009;16(10):296–299.

25.

Puri

Chaudry

Tatk

. N-hexane neuropathy in screen printers. Electromyogr Clin Neurophysiol. 2007;47(4):145–152.

26.

Misra

Kalita

. Toxic neuropathies. Neurol India. 2009;57(2):697–705.

27.

Kim

Kang

. Occupational neurological disorders in Korea. J Korean Med Sci. 2010;25(suppl):S26–S35.

28.

Takeuchi

. n-Hexane polyneuropathy in Japan: a review of n-hexane poisoning and its preventive measures. Environ Res. 1993;62(5):76–80.

29.

Zhang

Wang

Sun

. Pathology and neurophysiology analysis for peripheral neuropathy of four patients with chemicals poisoning [in Chinese]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi. 2012;30(4):126–129; discussion 129-130.

30.

Zhang

Wang

, et al. Misdiagnosis of occupational chronic n-hexane poisoning: an analysis of 16 cases [in Chinese]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi. 2014;32(3):930–931.

31.

Pan

Peng

Dai

Wang

Chuang

n-Hexane intoxication in a Chinese medicine pharmaceutical plant: a case report. J Med Case Rep. 2017;11(1):120. doi:10.1186/s13256-017-1280-9.

32.

Wang

Chen

Wang

. Electrophysiological follow-up of patients with chronic peripheral neuropathy induced by occupational intoxication with n-hexane. Cell Biochem Biophys. 2014;70(2):579–585. doi:10.1007/s12013-014-9959-7.

33.

Wade

. Apple releases new supplier responsibility report. 2001. https://www.business-humanrights.org/en/apple-releases-supply-chain-report-details-findings-on-child-labour-worker-illness-from-n-hexane-remedial-steps-pledges-greater-cooperation-with-chinese-ngos. See also: https://www.pbs.org/newshour/show/in-china-workers-detail-health-effects-of-producing-i-phones. Accessed February 6, 2019.

34.

Cheng

Shunchun

Wang

. Effect of integrated Chinese and western medicine in the treatment of peripheral neuropathy caused by n-hexane poisoning [in Chinese]. Clin J Chin Med. 2018;24. ISSN: 1674-7860. http://new.oversea.cnki.net/KCMS/detail/detail.aspx?dbcode=CJFQ&dbname=CJFDLAST2018&filename=ZYLY201824014&v=MDAwNDFyQ1VSTE9mWXVSbUZ5amdVYjNJUHpUSGQ3RzRIOW5PcTQ5RVlJUjhlWDFMdXhZUzdEaDFUM3FUcldNMUY=. Accessed February 6, 2019.

35.

Investigation of a chronic n-hexane poisoning accident in an electronic enterprise of Shenzhen City in 2017 [in Chinese]. Occup Health. 2019;5. doi:10.13329/j.cnki.zyyjk.2019.0187. http://new.oversea.cnki.net/KCMS/detail/detail.aspx?dbcode=CJFQ&dbname=CJFDLAST2019&filename=ZYJK201905029&v=MDAwNDFyQ1VSTE9mWXVSbUZ5amdWcjdBUHpUQlpiRzRIOWpNcW85SGJZUjhlWDFMdXhZUzdEaDFUM3FUcldNMUY=.Accessed February 6, 2019.

36.

Krasavage

O’Donoghue

DiVincenzo

Terhaar

. The relative neurotoxicity of methyl-n-butyl ketone, n-hexane and their metabolites. Toxicol Appl Pharmacol. 1980;52(3):433–441.

37.

Couri

Milks

. Toxicity and metabolism of the neurotoxic hexacarbons n-hexane, 2-hexanone, and 2,5-hexanedione. Annu Rev Pharmacol Toxicol. 1982;22(1):145–166.

38.

Iba

Fung

Gonzalez

. Functional Cyp2e1 is required for substantial in vivo formation of 2,5-hexanedione from n-hexane in the mouse. Arch Toxicol. 2000;74(2):582–586.

39.

Kim

, et al. Cytotoxicity of 1,2-diacetylbenzene in human neuroblastoma SHSY5Y cells is mediated by oxidative stress. Toxicology. 2008;243(1-2):216–223.

40.

Genter St Clair

Amarnath

Moody

Anthony

Anderson

Graham

. Pyrrole oxidation and protein cross-linking as necessary steps in the development of γ-diketone neuropathy. Chem Res Toxicol. 1988;1(3):179–185.

41.

DeCaprio

Kinney

LoPachin

. Comparative covalent protein binding of 2,5-hexanedione and 3-acetyl-2,5-hexanedione in the rat. Toxicol Environ Health A. 2009;72(2):861–869.

42.

Yin

Guo

Zeng

Zhao

Xie

. Correlation between levels of 2,5-hexanedione and pyrrole adducts in tissues of rats exposure to n-hexane for 5-days. PLoS One. 2013;8(9):e76011. doi:10.1371/journal.pone.0076011.

43.

Sayre

Shearson

Wongmongkolrit

Medori

Gambetti

. Structural basis of gamma-diketone neurotoxicity: non-neurotoxicity of 3,3-dimethyl-2,5-hexanedione, a gamma-diketone incapable of pyrrole formation. Toxicol Appl Pharmacol. 1986;84(1):36–44.

44.

Genter

Szakál-Quin

Anderson

Anthony

Graham

. Evidence that pyrrole formation is a pathogenetic step in gamma-diketone neuropathy. Toxicol Appl Pharmacol. 1987;87(2):351–362.

45.

DeCaprio

Briggs

Jackowski

Kim

. Comparative neurotoxicity and pyrrole-forming potential of 2,5-hexanedione and perdeuterio-2,5-hexanedione in the rat. Toxicol Appl Pharmacol. 1988;92(1):75–85.

46.

Gagnaire

Ensminger

Marignac

De Ceaurri

. Possible involvement of 1,2-diacetylbenzene in diethylbenzene-induced neuropathy in rats. J Appl Toxicol. 1991;11(2):261–268.

47.

Gagnaire

Marignac

De Ceaurriz

. Diethylbenzene-induced sensorimotor neuropathy in rats. J Appl Toxicol. 1990;10(8):105–112.

48.

Zhan

C-G

Spencer

Dixon

. Computational insights into the chemical structures and mechanisms of the chromogenic and neurotoxic effects of aromatic γ-diketones. J Phys Chem B. 2003;107(6):2853–2861.

49.

Kim

M-S

Hashemi

Spencer

Sabri

. Amino acid and protein targets of 1,2-diacetylbenzene, a potent aromatic γ-diketone that induces proximal neurofilamentous axonopathy. Toxicol Appl Pharmacol. 2002;18355–18365.

50.

Kim

Sabri

Miller

Kayton

Dixon

Spencer

. 1,2-Diacetylbenzene, the neurotoxic metabolite of a chromogenic aromatic solvent, induces proximal axonopathy. Toxicol Appl Pharmacol. 2001;177(2):121–131.

51.

Tshala-Katumbay

Monterroso

Kayton

Lasarev

Sabri

Spencer

. Probing mechanisms of axonopathy. Part I: protein targets of 1,2-diacetylbenzene, the neurotoxic metabolite of aromatic solvent 1,2-diethylbenzene. Toxicol Sci. 2008;105(3):134–141.

52.

Tshala-Katumbay

Palmer

Kayton

Sabri

Spencer

. A new murine model of giant proximal axonopathy. Acta Neuropathol. 2005;109(1):405–410.

53.

Tshala-Katumbay

Palmer

Lasarev

Kayton

Sabri

Spencer

. Monocyclic and dicyclic hydrocarbons: structural requirements for proximal giant axonopathy. Acta Neuropathol. 2006;112(2):17–24.

54.

Kim

Park

, et al. Neurotoxic effect of 2,5-hexanedione on neural progenitor cells and hippocampal neurogenesis. Toxicology. 2009;260(1):97–103.

55.

Cheng

Wang

, et al. Exposure to 2,5-hexanedione can induce neural malformations in chick embryos. Neurotoxicology. 2012;33(4):1239–1247.

56.

Anthony

Boejelheide

Graham

. The effect of 3,4-dimethyl substitution on the neurotoxicity of 2,5-hexanedione. I. Accelerated clinical neuropathy is accompanied by more proximal axonal swellings. Toxicol Appl Pharmacol. 1983;71(2):362–372.

57.

Spencer

Sterman

Horoupian

Foulds

. Neurotoxic fragrance produces ceroid and myelin disease. Science. 1979;204(5):633–635.

58.

Spencer

Sterman

Horoupian

Bischoff

Foster

. Neurotoxic changes in rats exposed to the fragrance compound acetyl ethyl tetramethyl tetralin. Neurotoxicology. 1979;1(5):221–237.

59.

Palmer

Tshala-Katumbay

Hashemi

Sabri

Spencer

. Tetralin and metabolites: protein reactivity, chromogenicity and neurotoxicity. Toxicol Sci. 2004;78(5):S–1. Abstract.

60.

Gerarde

. Toxicology and biochemistry of aromatic hydrocarbons. In: Browning

, ed. Elsevier Monographs on Toxic Agents. Amsterdam, the Netherlands: Elsevier; 1960:73.

61.

American Chemistry Council, Ethyl Benzene Panel HPV Task Group, Diethylbenzene Subteam. Report for the diethylbenzene-rich streams category. High Production Volume (HPV) Chemical Challenge Program. 2004. https://www.americanchemistry.com/ProductsTechnology/Product-Specific-Groups/. Accessed February 6, 2019.

62.

Spencer

Schaumburg

. Ultrastructural studies of the dying-back process. III. The evolution of experimental peripheral giant axonal degeneration. J Neuropathol Exp Neurol. 1977;36(1):2762–2799.

63.

Schaumburg

Spencer

. Environmental hydrocarbons produce degeneration in cat hypothalamus and optic tract. Science. 1978;199(4325):199–200.

64.

Jones

Cavanagh

. Recovery from 2,5-hexanediol intoxication of the retinotectal tract of the rat. An ultrastructural study. Acta Neuropathol. 1982;58:286–290.

65.

Veronesi

Peterson

Bornstein

Spencer

. Ultrastructural studies of the dying-back process. VI. Examination of nerve fibers undergoing giant axonal degeneration in organotypic culture. J Neuropathol Exp Neurol. 1983;42(1):153–165.

66.

Anthony

Giangaspero

Graham

. The spatio-temporal pattern of the axonopathy associated with the neurotoxicity of 3,4-dimethyl-2,5-hexanedione in the rat. J Neuropathol Exp Neurol. 1983;42(5):548–660.

67.

Lehning

Jortner

Fox

Arezzo

Kitano

LoPachin

. Gamma-diketone peripheral neuropathy. I. Quality morphometric analyses of axonal atrophy and swelling. Toxicol Appl Pharmacol. 2000;165(5):127–140.

68.

LoPachin

Jortner

Reid

Das

. Gamma-diketone central neuropathy: quantitative morphometric analysis of axons in rat spinal cord white matter regions and nerve roots. Toxicol Appl Pharmacol. 2003;193(6):29–46.

69.

Marszalek

Lee

, et al. Subunit composition of neurofilaments specifies axonal diameter. J Cell Biol. 1996;133(2):1061–1069.

70.

Zagoren

Politis

Spencer

. Rapid reorganization of the axonal cytoskeleton induced by a γ-diketone. Brain Res. 1983;270(1):162–164.

71.

Spencer

Schaumburg

. Central-peripheral distal axonopathy—the pathology of dying-back polyneuropathies. In: Zimmerman

, ed. Progress in Neuropathology. New York, NY: Grune and Stratton; 1976:253–295.

72.

Spencer

Thomas

. Ultrastructural studies of the dying-back process. II. The sequestration and removal by Schwann cells and oligodendrocytes of organelles from normal and diseases axons. J Neurocytol. 1974;3(2):763–783.

73.

Politis

Pellegrino

Spencer

. Ultrastructural studies of the dying-back process. V. Axonal neurofilaments accumulate at sites of 2,5-hexanedione application: evidence for nerve fibre dysfunction in experimental hexacarbon neuropathy. J Neurocytol. 1980;9(1):505–516.

74.

Graham

Anthony

Boekelheide

, et al. Studies of the molecular pathogenesis of hexane neuropathy. II. Evidence that pyrrole derivatization of lysyl residues leads to protein crosslinking. Toxicol Appl Pharmacol. 1982;64(6):415–422.

75.

Lapadula

Irwin

Suwita

Abou-Donia

. Cross-linking of neurofilament proteins of rat spinal cord in vivo after administration of 2,5-hexanedione. J Neurochem. 1986;46(1):1843–1850.

76.

Boekelheide

. 2,5-Hexanedione alters microtubule assembly. II. Enhanced polymerization of crosslinked tubulin. Toxicol Appl Pharmacol. 1987;88(3):383–396.

77.

Dixit

Ross

Goldman

Holzbaur

ELF

. Differential regulation of dynein and kinesin motor proteins by tau. Science. 2008;319(5866):108–109.

78.

Sabri

Hashemi

Lasarev

Spencer

. Axonopathy-inducing 1,2-diacetylbenzene forms adducts with motor and cytoskeletal proteins required for axonal transport. Neurochem Res. 2007;32(12):2152–2159.

79.

Opanashuk

Lehning

LoPachin

. Gamma-diketone peripheral neuropathy. III. Neurofilament gene expression. Neurotoxicology. 2001;22(2):215–220.

80.

Wang

Qiu

Gao

, et al. 2,5-Hexanedione downregulates nerve growth factor and induces neuron apoptosis in the spinal cord of rats via inhibition of the PI3K/Akt signaling pathway. PLoS One. 2017;2(6):e0179388. doi:10.1371/journal.pone.0179388.

81.

Wang

Liu

, et al. 2,5-Hexanedione increases the percentage of proliferative Sox2⁺ cells in rat hippocampus. Toxicol Ind Health. 2018;34(9):589–595. doi:10.1177/0748233718772767.

82.

Liu

Sun

, et al. N-hexane inhalation during pregnancy alters DNA promoter methylation in the ovarian granulosa cells of rat offspring. J Appl Toxicol. 2014;34:841–856.

83.

Sun

Lin

Liu

Sheng

Zheng

. 2,5-Hexanedione induces human ovarian granulosa cell apoptosis through BCL-2, BAX, and CASPASE-3 signaling pathways. Arch Toxicol. 2012;86(2):205–215. doi:10.1007/s00204-011-0745-7.

84.

Liu

Huang

Pang

Zhang

. The effect of n-hexane on the gonad toxicity of female mice. Biomed Environ Sci. 2012;25(2):189–196.

85.

Mohan

John

. Microtubule-associated proteins as direct crosslinkers of actin filaments and microtubules. UBMB Life. 2015;67(1):395–403.

86.

Fung

Kettenbach

Chakrabarti

Higgs

. A complex containing lysine-acetylated actin inhibits the formin INF2. Nat Cell Biol. 2019;21(6):592–602. doi:10.1038/s41556-019-0307-4.

87.

Bis-Brewer

Danzi

Wuchty

Züchner

. A network biology approach to unraveling inherited axonopathies. Sci Rep. 2019;9(2):1692. doi:10.1038/s41598-018-37119-z.

88.

Henty-Ridilla

Rankova

Eskin

Kenny

Goode

. Accelerated actin filament polymerization from microtubule plus ends. Science. 2016;352(4):1004–1009. doi:10.1126/science.aaf1709.

89.

Lim

Ito

Sakata-Sogawa

Tokunaga

. CLIP-170 is essential for MTOC repositioning during T cell activation by regulating dynein localisation on the cell surface. Sci Rep. 2018;8(1):17447. doi:10.1038/s41598-018-35593-z.

90.

Toba

Jin

Yamada

, et al. Alpha-synuclein facilitates to form short unconventional microtubules that have a unique function in the axonal transport. Sci Rep. 2017;7(2):16386. doi:10.1038/s41598-017-15575-3. Correction: doi:10.1038/s41598-018-25979-4.

91.

Ozon

Maucuer

Sobel

. The stathmin family—molecular and biological characterization of novel mammalian proteins expressed in the nervous system. Eur J Biochem. 1997;248(3):794–806.

92.

Horwitz

Shen

, et al. The microtubule-destabilizing activity of metablastin (p19) is controlled by phosphorylation. J Biol Chem. 1997;272(1):8129–8132.

93.

Liedtke

Leman

Fyff

Raine

Schubart

. Stathmin-deficient mice develop an age-dependent axonopathy of the central and peripheral nervous systems. Am J Pathol. 2002;160(6):469–480.

94.

Sabri

Hashemi

Chohan

, et al. γ-Diketone toxicity: a role for stathmin in nerve and testes damage? Presented at: Pacific Northwest Association of Toxicologists, 21st Annual Meeting. Bend, OR; 2004.

95.

Kassa

Monterroso

Wentzell

, et al. Proximal giant neurofilamentous axonopathy in mice genetically engineered to resist calpain and caspase cleavage of α-II spectrin. J Mol Neurosci. 2012;47(5):631–638. doi:10.1007/s12031-011-9699-8.

96.

Spencer

Sabri

Schaumburg

Moore

. Does a defect of energy metabolism in the nerve fiber underlie axonal degeneration in polyneuropathies? Ann Neurol. 1979;5(2):501–507.

97.

Sickles

Fowler

Testino

. Effects of neurofilamentous axonopathy-producing neurotoxicants on in vitro production of ATP by brain mitochondria. Brain Res. 1990;528(1):25–31.

98.

Sabri

. In vitro effect of n-hexane and its metabolites on selected enzymes in glycolysis, pentose phosphate pathway and citric acid cycle. Brain Res. 1984;297(3):145–150.

99.

Adedara

Abolaji

Odion

, et al. Impairment of hepatic and renal functions by 2,5-hexanedione is accompanied by oxidative stress in rats. J Toxicol. 2014;2014(5):239240. doi:10.1155/2014/239240.

100.

Kang

Kim

. Oxidative stress with tau hyperphosphorylation in memory impaired 1,2-diacetylbenzene-treated mice. Tox Lett. 2017;279(3):53–59. doi:10.1016/j.toxlet.2017.07.892.

101.

Sakai

Araki

Ushio

Takeuchi

Ikeya

. Effect of hydrolysis conditions on the determination of urinary 2,5-hexanedione in workers exposed or not exposed to N-hexane [in Japanese]. Sangyo Igaku. 1992;34(2):440–447.

102.

Perbellini

Pezzoli

Brugnone

Canesi

. Biochemical and physiological aspects of 2,5-hexanedione: endogenous or exogenous product? Int Arch Occup Environ Health. 1993;65(1):49–52.

103.

Chambers

Blount

McElprang

Waterhouse

Morrow

. Picogram measurement of volatile n-alkanes (n-hexane through n-dodecane) in blood using solid-phase microextraction to assess nonoccupational petroleum-based fuel exposure. Anal Chem. 2008;80(5):4666–4674.

104.

Bavazzano

Apostoli

Balducci

, et al. Determination of urinary 2,5-hexanedione in the general Italian population. Int Arch Occup Environ Health. 1988;71(2):284–288.

105.

Xing-Fu

Ya-Ling

Wei

, et al. Determination of total urinary 2,5-hexanedione in the Chinese general population. Environ Res. 2016;150(4):645–650. doi:10.1016/j.envres.2016.05.030.

106.

Wang

Zhou

Wang

Yin

Xie

. Determination of normal reference value of pyrrole adducts in urine in young people in a university in Shandong, China [in Chinese]. Zhonghua Lao Dong Wei Sheng Zhi Ye Bing Za Zhi. 2015;33(1):435–437.

107.

Zlatkis

Poole

Brazeli

Bafus

Spencer

. Volatile metabolites in sera of normal and diabetic subjects. J Chromatogr. 1980;182(2):137–145.

108.

Altenkirch

Stoltenburg

Wagner

. Experimental studies on hydrocarbon neuropathies induced by methyl-ethyl-ketone (MEK). J Neurol. 1978;219(3):159–170.

109.

Veronesi

Lington

Spencer

. A tissue culture model of methyl ethyl ketone’s potentiation of n-hexane neurotoxicity. Neurotoxicology. 1984;5(4):43–52.

110.

Misumi

Nagano

. Experimental study on the enhancement of the neurotoxicity of methyl n-butyl ketone by non-neurotoxic aliphatic monoketones. Br J Ind Med. 1985;42(1):155–161.

111.

Hattis

Landaw

Froines

. Toxicokinetic interaction of 2,5-hexanedione and methyl ethyl ketone. Arch Toxicol. 2002;75(2):643–652.

112.

Spencer

Ochoa

. The mammalian peripheral nervous system in old age. In: Johnson

Jr , ed. Aging and Cell Structure. New York, NY: Plenum; 1981:35–103.

113.

Spencer

Bischoff

Schaumburg

. On the specific molecular configuration of neurotoxic aliphatic hexacarbon compounds causing central-peripheral distal axonopathy. Toxicol Appl Pharmacol. 1976;44(1):17–28.

114.

DeCaprio

Strominger

Weber

. Neurotoxicity and protein binding of 2,5-hexanedione in the hen. Toxicol Appl Pharmacol. 1983;68(3):297–307.

115.

Markelewicz

Jr Hall

Boekelheide

. 2,5-Hexanedione and carbendazim coexposure synergistically disrupts rat spermatogenesis despite opposing molecular effects on microtubules. Toxicol Sci. 2004;80(3):92–100.

116.

Liu

Duan

Dai

, et al. The effect of 2,5-hexanedione on permeability of blood-nerve barrier in rat. Hum Exp Toxicol. 2010;29(1):497–506.

117.

Chen

Kanai

Cowan

Hirokawa

. Projection domains of MAP2 and tau determine spacings between microtubules in dendrites and axons. Nature. 1992;360(9):674–677.

118.

Gotow

. Neurofilament cross-bridge—a structure associated specifically with the neurofilament among the intermediate filament family. In: Nixon

Yuan

, eds. Advances in Neurobiology 3, Cytoskeleton of the Nervous System; 2010:225–247.

119.

Gotow

Tanaka

. Phosphorylation of neurofilament H subunit as related to arrangement of neurofilaments. J Neurosci Res. 1994;37(6):691–713.

120.

Chen

Nakata

Zhang

Hirokawa

. The C-terminal tail domain of neurofilament protein-H (NF-H) forms the crossbridges and regulates neurofilament bundle formation. J Cell Biol. 2000:113(1):3861–3869.

121.

Barry

Carpenter

Yager

, et al. Variation of the neurofilament medium KSP repeat sub-domain across mammalian species: implications for altering axonal structure. J Exp Biol. 2010;213(6):128–136. doi:10.1242/jeb.033787.

122.

Zhu

Lindenbaum

Levavasseur

Jacomy

Julien

. Disruption of the NF-H gene increases axonal microtubule content and velocity of neurofilament transport: relief of axonopathy resulting from the toxin beta, beta’-iminodipropionitrile. J Cell Biol. 1988;143(1):183–193.

123.

Jung

Chylinski

Pimenta

Ortiz

Shea

. Neurofilament transport is dependent on actin and myosin. J Neurosci. 2004;24(4):9486–9496.

124.

Robert

Herrmann

Davidson

Gelfand

. Microtubule-dependent transport of vimentin filament precursors is regulated by actin and by the concerted action of Rho- and p21-activated kinases. FASEB J. 2014;28(3):2879–2890. doi:10.1096/fj.14-250019.

125.

Hitchcock-De Gregori

Mandala

Sachs

. Changes in actin lysine reactivities during polymerization detected using a competitive labeling method. J Biol Chem. 1982;257(1):12573–12580.

126.

Charbaut

Curmi

Ozon

Lachkar

Redeker

Sobel

. Stathmin family proteins display specific molecular and tubulin binding properties. J Biol Chem. 2001;276(2):16146–16154.

127.

Nouar

Breuzard

Bastonero

, et al. Direct evidence for the interaction of stathmin along the length and the plus end of microtubules in cells. FASEB J. 2016;30(9):3202–3215. doi:10.1096/fj.201500125R.

Neuroprotein Targets of γ-Diketone Metabolites of Aliphatic and Aromatic Solvents That Induce Central–Peripheral Axonopathy

Abstract

Keywords

Introduction

γ-Diketone-Induced Changes in Brain Gene Expression

γ-Diketone-Induced Changes in Neuroprotein Abundance

Early CNS Protein Changes

Later CNS Protein Changes

Normal Aging and Diabetes Mellitus

Conclusion

Footnotes

Acknowledgments

Declaration of Conflicting Interests

Funding

ORCID iD

References