Nervous system degeneration produced by the industrial solvent methyl n-butyl ketone

RIFM fragrance ingredient safety assessment, 2-decanone, CAS Registry Number 693-54-9

The existing information supports the use of this material as described in this safety assessment. 2-Decanone was evaluated for genotoxicity, repeated dose toxicity, reproductive toxicity, local respiratory toxicity, phototoxicity/photoallergenicity, skin sensitization, and environmental safety. Data from read-across analog 2-heptanone (CAS # 110-43-0) show that 2-decanone is not expected to be genotoxic. Data on read-across analog 2-heptanone (CAS # 110-43-0) provide a calculated margin of exposure (MOE) > 100 for the repeated dose toxicity and reproductive toxicity endpoints. The skin sensitization endpoint was completed using the dermal sensitization threshold (DST) for non-reactive materials (900 μg/cm²); exposure is below the DST. The phototoxicity/photoallergenicity endpoints were evaluated based on ultraviolet (UV) spectra; 2-decanone is not expected to be phototoxic/photoallergenic. For the local respiratory endpoint, a calculated MOE >100 was provided by the read-across analog 4-methyl-2-pentanone (CAS # 108-10-1). The environmental endpoints were evaluated; for the hazard assessment based on the screening data, 2-decanone is not persistent, bioaccumulative, and toxic (PBT) as per the International Fragrance Association (IFRA) Environmental Standards. For the risk assessment, 2-decanone was not able to be risk screened as there were no reported volumes of use for either North America or Europe in the 2015 IFRA Survey.

Clinical description of toxic neuropathies

Toxic neuropathy, although rare, is an important consideration in the setting of a known or suspected toxic exposure in the workplace or other environment. This chapter discusses the clinical and electrodiagnostic evaluation of peripheral neuropathies, highlighting findings that direct further workup and may point to specific toxins as etiology. The difficulty of establishing causality of a toxin in relation to peripheral neuropathy is discussed; guidelines for establishing causality are presented. Examples of common industrial toxins are listed, including their typical industrial uses and their mechanisms of action in producing neuropathy. Characteristic clinical presentations of specific toxic neuropathies are highlighted with selected case studies.

Are sperm parameters the best indicator of 2,5-hexanedione reproductive toxicity in male rats?

In this study on 40 male Wistar rats of approximately 3-weeks-old, the effects of 2,5-hexanedione on sperm principal parameters were examined by subcutaneously administering 100, 200 and 400 mg/(kgday) for 12 weeks. Other indicators of 2,5-hexanedione intoxication were examined: testes, epididymides, body weight, limb paralysis and histology. The sperm motility was adversely affected in all of the three treatment groups. No sperm motility at all was observed in the 200 and 400 mg/(kgday) treatment groups and rapid progressive motility was significantly reduced in the 100 mg/(kgday) group (P<0.05). The sperm concentration and morphology, testes, epididymides and body weight were significantly affected in the 200 and 400 mg/(kgday) groups but not in the 100 mg/(kgday) group (P<0.05). All the rats in the 400 mg/(kgday) group could not move due to leg paralysis, whereas only 50% of the rats in the 200mg/(kgday) group and none in the 100 mg/(kgday) were affected. Histological examination of the testes in the 400, 200 and 100 mg/(kgday) groups revealed Sertoli cell-only syndrome in 70, 30 and 0% of the rats, respectively. In conclusion, this study has shown that sperm motility was the most significantly sensitive indicator of 2,5-hexanedione intoxication in all the treatment groups.

γ-Diketone neuropathy: axon atrophy and the role of cytoskeletal protein adduction

Multifocal giant neurofilamentous axonal swellings and secondary distal degeneration have been historically considered the hallmark features of gamma-diketone neuropathy. Accordingly, research conducted over the past 25 years has been directed toward discerning mechanisms of axonal swelling. However, this neuropathological convention has been challenged by recent observations that swollen axons were an exclusive product of long-term 2.5-hexanedione (HD) intoxication at lower daily dose-rates (e.g., 175 mg/kg/day); that is, higher HD dose-rates (e.g., 400 mg/kg/day) produced neurological deficits in the absence of axonal swellings. The observation that neurological toxicity can be expressed without axonal swelling suggests that this lesion is not an important pathophysiological event. Instead, several research groups have now shown that axon atrophy is prevalent in nervous tissues of laboratory animals intoxicated over a wide range of HD dose-rates. The well-documented nerve conduction defects associated with axon atrophy, in conjunction with the temporal correspondence between this lesion and the onset of neurological deficits, strongly suggest that atrophy has pathophysiological significance. In this commentary, we present evidence that supports a pathognomonic role for axon atrophy in gamma-diketone neuropathy and suggests that the functional consequences of this lesion mediate the corresponding neurological toxicity. Previous research has demonstrated that HD interacts with proteins via formation of pyrrole adducts. We therefore discuss the possibility that this chemical process is essential to the mechanism of atrophy. Evidence presented in this review suggests that "distal axonopathy" is an inaccurate classification and future nosological schemes should be based on the apparent primacy of axon atrophy.

Safety assessment of MIBK (methyl isobutyl ketone)

MIBK (Methyl Isobutyl Ketone) is an aliphatic ketone that functions as both a denaturant and solvent in cosmetic products. Current use in cosmetic products is very limited, but MIBK is reported to be used in one nail correction pen (volume = 3 ml) at a concentration of 21%. The maximum percutaneous absorption rate in guinea pigs is 1.1 micromol/min/cm2 at 10 to 45 min. Metabolites include 4-hydroxy-4-methyl-2-pentanone (oxidation product) and 4-methyl-2-pentanol (4-MPOL) (reduction product). Values for the serum half-life and total clearance time of MIBK in animals were 66 min and 6 h, respectively. In clinical tests, most of the absorbed MIBK had been eliminated from the body 90 min post exposure. MIBK was not toxic via the oral or dermal route of exposure in acute, short-term, or subchronic animal studies, except that nephrotoxicity was observed in rats dosed with 1 g/kg in a short-term study. MIBK was an ocular and skin irritant in animal tests. Ocular irritation was noted in 12 volunteers exposed to 200 ppm MIBK for 15 min in a clinical test. A depression of the vestibulo-oculomotor reflex was seen with intravenous infusion of MIBK (in an emulsion) at 30 microM/kg/min in female rats. The no-observed-effect level in rats exposed orally to MIBK was 50 mg/kg. Both gross and microscopic evidence of lung damage were reported in acute inhalation toxicity studies in animals. Short-term and subchronic inhalation exposures (as low as 100 ppm) produced effects in the kidney and liver that were species and sex dependent. Dermal doses of 300 or 600 mg/kg for 4 months in rats produced reduced mitotic activity in hair follicles, increased thickness of horny and granular cell layers of the epidermis, a decrease in the number of reactive centers in follicles (spleen), an increase in the number of iron-containing pigments in the area of the red pulp (spleen), and a reduction in the lipid content of the cortical layer of the adrenal glands. Neuropathological changes in the most distal portions of the tibial and ulnar nerves were observed in young adult rats which inhaled 1500 ppm MIBK for up to 5 months. No adverse effects were seen in any other neurological end point by any route of exposure in other studies using rats or other animal species. Clinical tests demonstrated a threshold for MIBK-induced irritation of the lungs at 0.03 to 0.1 mg/L after 1 min of respiration. MIBK was not mutagenic in the Ames test or in a mitotic gene-conversion assay in bacteria. Mammalian mutagenicity test results were also negative in the following assays: mouse lymphoma, unscheduled DNA synthesis, micronucleus, cell transformation, and chromosome damage. MIBK did not induce any treatment-related increases in embryotoxicity or fetal malformations in pregnant Fischer 344 rats or CD-1 mice that inhaled MIBK at concentrations of 300, 1000, or 3000 ppm. There was evidence of treatment-related maternal toxicity only at the highest concentration tested. MIBK applied to the tail of rats daily at doses of 300 or 600 mg/kg for 4 months produced changes in the testes, including a reduction in the number of spermatocytes, spermatids, and spermatozoa. An ongoing carcinogenicity study of MIBK being conducted by the National Toxicology Program will be considered when the results are available. On the basis of the information that is currently available, MIBK is considered safe as used in nail polish removers and as an alcohol denaturant in cosmetic products.

Spectrophotometric analysis of solubilized rat hair proteins following intraperitoneal injection of 2,5-hexanedione

The neurotoxic industrial solvents n-hexane and methyl n-butyl ketone are toxic by virtue of their common metabolite, 2,5-hexanedione (2,5-HD). Our previous work showed that pyrrole-like substances in solubilized rat hair proteins from rats injected (ip) daily with 2,5-HD demonstrated maximal absorbance in the 530-nm spectral region following reaction with Ehrlich's reagent (p-dimethylaminobenzaldehyde). Modification of the current analytical methods of achieving high specificity and lower detection limits with small sample quantities could have important implications for monitoring human populations. Adult male Sprague-Dawley rats were housed in individual metabolic cages with food and water provided ad libitum. Individual rats were injected (ip) daily with either 50 mg/kg 2,5-HD or physiologic-buffered saline (PBS). Plucked hair samples (dorsal, right flank, and left flank) were obtained from each rat before and at 7-day intervals after exposure to 2,5-HD or PBS for 28 days. Hair proteins solubilization and extraction procedures were adapted from earlier studies. We read 1 mL of dialyzed hair protein solution (2,5-HD or PBS control) against a reference cuvette containing water. Analyses utilized a Shimadzu UV 160 V recording spectrophotometer at an absorbency spectral range of 450 to 600 nm. In all spectral tracings, absorbance maxima (at 530 nm) characteristic of pyrrole-like substances were detected only in samples from 2,5-HD-treated rats. Absorbance at 530 nm was detected starting at Day 7 after exposure. The authors acknowledge Dr. Richard Whorton and Dr. Barbara Buckley for advice and for the use of their spectrophotometric equipment and Dr. Lowell A. Goldsmith for his help in our choice of the subject studied. This work was supported by the Walter P. Inman Memorial Fund in an award from Duke University to Dr. Leon Lack.

Controversies in neurotoxicology: current status

Several controversial neurotoxic syndromes have received notoriety in the past several decades. For each, the controversy involves the most fundamental question about the existence of the disorder as a clinically diagnosable entity. Interestingly, the most controversial of these syndromes share several features, including argument about the existence of each syndrome in the courtroom. The authors focus their discussion on three problems (painter's encephalopathy, silicone breast implant neurotoxicity, and the Gulf War syndrome) for which no scientific consensus has been reached that would establish them as diagnosable disorders. These syndromes do not meet traditional disease criteria, and until a clear set of symptoms and objective signs can be defined, a definite course and clear cause demonstrated, and specific tests and treatments identified, these syndromes are likely to remain highly controversial.

Neurologic evaluation of workers previously diagnosed with solvent-induced toxic encephalopathy

We examined 52 railroad workers with long-term occupational solvent exposures (average 22 years duration) who had been previously diagnosed by others as having solvent-induced toxic encephalopathy. All described episodes of transient intoxication associated with occupational solvent exposure. Persistent symptoms developed, an average, 16 years after exposure onset and included impaired memory (38), altered mood (21), imbalance (18), and headache (17). Thirteen workers had mild mental status abnormalities, but none fulfilled conventional clinical criteria for encephalopathy or dementia. None had abnormal blink reflex (51) or abnormal electroencephalographic (39) studies. Eight of 47 magnetic resonance imaging studies showed evidence of scattered ischemic lesions among workers with known diabetes mellitus (2), elevated blood pressure (4), or peripheral vascular disease (2). One magnetic resonance imaging scan showed mild cortical atrophy. In stepwise multiple linear and logistic regression models, no statistically significant (P < 0.05) dose-response relationships were found between exposure duration and symptoms or signs that were suggestive of encephalopathy. However, the number of symptoms (P < 0.001) and the number of signs (P = 0.05) were associated with current use of central nervous system-active medications. Further, lower Mini-Mental Status Examination scores were associated with a history of alcohol abuse (P = 0.01) and lower educational level (P = 0.03). The number of chief symptoms involving memory, mood, balance, or headache differed significantly among workers in different geographic sites (F(3.48) = 2.94, P = 0.04), a finding that was not explained by job title or exposure duration. There also was a significant (P = 0.0001) inverse relationship between initial exposure year (r2 = 0.60) or total years of exposure through 1987 (r2 = 0.56) and interval to major neurologic symptom onset, suggesting that factors other than solvent exposure account in part for worker complaints. We found no objective neurologic evidence supportive of toxic encephalopathy or any other uniform syndrome among these individuals, and most complaints were explained by neuropsychological factors or conditions unrelated to occupational solvent exposure.

Axonal neurofilaments are nonessential elements of toxicant-induced reductions in fast axonal transport: video-enhanced differential interference microscopy in peripheral nervous system axons

Neurofilament modification and accumulation, occurring in toxicant-induced neuropathies, has been proposed to compromise fast axonal transport and contribute to neurological symptoms or pathology. The current study compares the effects of the neurotoxicants acrylamide (ACR) and 2,5-hexanedione (2,5-HD) on the quantity of fast, bidirectional vesicular traffic within isolated mouse sciatic nerve axons from transgenic mice lacking axonal neurofilaments (Eyer and Peterson, Neuron 12, 1-20, 1994) and nontransgenic littermates possessing neurofilaments. Fast anterograde and retrograde membrane bound organelle (MBO) traffic was quantitated within axons, before and after toxicant exposure, using video-enhanced differential interference contrast (AVEC-DIC) microscopy. Addition of 0.7 mM ACR to the buffer bathing the nerve produced a time-dependent reduction in bidirectional transport with a similar time to onset and magnitude in both transgenic and nontransgenic mice. 2,5-HD (4 mM) exposure reduced bidirectional vesicle traffic by a similar amount in both transgenic and nontransgenic animals. The time to onset of the transport reduction was less and the magnitude of the reduction was greater with 2,5-HD compared to ACR. A single 10-min exposure to ACR or 2,5-HD produced a similar reduction in transport to that produced by prolonged (1 h) exposure. Nonneurotoxic propionamide or 3,4-hexanedione (3,4-HD) produced no changes in bidirectional transport in either transgenic or nontransgenic animals. We conclude that ACR or 2,5-HD produces a rapid, saturable, nonreversible, neurotoxicant-specific reduction in fast bidirectional transport within isolated peripheral nerve axons. These actions are mediated through direct modification of axonal component(s), which are independent of toxicant-induced modifications of, or accumulations of, neurofilaments.

Absence of polyneuropathy among workers previously diagnosed with solvent-induced toxic encephalopathy

An association between polyneuropathy and occupational exposure to trichloroethylene, trichloroethane, perchloroethylene, or similar solvents alone or in combination is controversial. We sought to determine whether workers previously diagnosed with solvent-induced toxic encephalopathy had objective evidence of polyneuropathy. Thirty railroad workers previously diagnosed with toxic encephalopathy were examined in the context of litigation against their employers. All described long-term occupational solvent exposure averaging 20 years in duration (range, 10 to 29 years) and producing acute intoxication on a regular basis. The diagnosis of subclinical or clinical polyneuropathy was established using a combination of symptoms, signs, and nerve conduction study (NCS) measures, consistent with standard clinical practice. Potential confounders were identified. NCS results were compared with historical controls, including unexposed workers matched by gender, age, and body mass index. Dose-response relationships were evaluated using simple linear and stepwise regression models. Three workers fulfilled clinical polyneuropathy criteria. The only worker fulfilling NCS criteria for confirmed clinical polyneuropathy had diabetes mellitus. Mean NCS values for most measures were similar to control values, and existing differences in sensory amplitudes disappeared when compared with the matched control group. NCS measures were not significantly influenced by exposure duration or job title. Separation into groups on the basis of the presence or absence of polyneuropathy symptoms, previous diagnosis of polyneuropathy, disability status, and severity or type of encephalopathy did not demonstrate significant NCS differences. The complaints of these workers claiming neurotoxic injury from occupational solvent exposure are not explained by peripheral nervous system dysfunction.

Protein-bound pyrroles in rat hair following subchronic intraperitoneal injections of 2,5-hexanedione

Studies were initiated to ascertain whether body hair could be used to develop a biological marker for chronic exposure to industrial neurotoxicants that yield the metabolite 2,5-hexanedione (2,5-HD), that is, n-hexane and methyl n-butyl ketone. Rats were injected daily with a 50 mg/kg ip dose of 2,5-HD for 45 d. At intervals, body hair and individual vibrissae were removed (under general anesthesia) and tested for the presence of pyrrole substances with p-N,N-dimethylaminobenzaldehyde (DMAB, Ehrlich's reagent). Vibrissae and body hair were stained a reddish color that was distinctly different from that observed with the hair taken from control animals. Solubilized body hair protein from the treated animals gave a positive Ehrlich's test, while that from control animals was negative. Spectral analysis of the DMAB-treated hair from experimental animals disclosed a maximum absorbance at 530 nm, which indicated the presence of pyrrole substituents. Serial analysis of individual nose hairs taken during 2,5-HD administration showed a progression with time of the region staining positively for pyrroles, thus indicating that the process can proceed in growing hair. These findings suggest the potential utility of hair as an indicator for chronic exposure to this class of industrial chemicals possessing neurotoxicity potential. This could complement urinary analysis, which is now used to confirm recent exposure.

In vitro binding of [14C]2,5-hexanedione to rat neuronal cytoskeletal proteins

2,5-Hexanedione (2,5-HD) induces central-peripheral axonpathy characterized by the accumulation of 10-nm neurofilaments proximal to the nodes of Ranvier and a Wallerian-type degeneration. It has been postulated that neurofilament crosslinking may be involved in the production of this axonopathy. A potential initiating event in this neurotoxic process may be the direct binding of 2,5-HD to neurofilament and microtubule proteins. In this study, the in vitro binding of [14C]2,5-HD to neurofilament and microtubule proteins was examined. Neurofilament proteins isolated from rat spinal cord or microtubule proteins isolated from rat brain were incubated in the presence of 2,5-HD at concentrations ranging from 25 to 500 mM. Quantitative analysis of sodium dodecyl sulfate (SDS) polyacrylamide gels revealed a dose- and time-dependent binding of 2,5-HD to both neurofilament proteins and microtubule proteins. Expressed as pmol 2,5-HD bound per microgram protein, the observed relative binding was MAP2 > NF160 > NF200 > > NF68 > tubulin. These data demonstrate the direct binding of 2,5-HD to cytoskeletal proteins including both neurofilaments and microtubules.

Giant axonopathy in streptozotocin diabetes of rats

The ultrastructure of peripheral sensory nerves was investigated in adult Wistar rats suffering from experimental diabetes mellitus 6 and 10 weeks after the injection of streptozotocin. Giant axons were seen in sections from the nerves of streptozotocin-treated rats; some contained masses of neurofilaments, others were predominantly filled with ill-defined vesicles. At the swollen axons, the myelin sheath was thinned or absent. In other regions, large intramyelinic vacuoles were observed. A number of nerve fibers broke down completely and underwent Wallerian degeneration. This was accompanied by Schwann cell proliferation and formation of Büngner bands. Concomitantly with axonal degeneration, nerve regeneration started from intact internodes. The pathomorphology of streptozotocin diabetic neuropathy closely resembles that of some toxic distal axonopathies. This points to a common metabolic basis of giant axonopathies of different etiology.

Peripheral neuropathy with giant axons and cardiomyopathy associated with desmin type intermediate filaments in skeletal muscle

A sporadic case (female, aged 14 years) is reported who was affected by myopathy, restrictive cardiomyopathy and sensory motor polyneuropathy. A muscle biopsy showed accumulation of osmiophilic granular and filamentous material on electron microscopy, which stained positively in immunofluorescence for desmin. Increased desmin phosphorylated isoforms have been demonstrated by one- and two-dimensional electrophoresis. Sural nerve biopsy showed a peripheral neuropathy with giant axons, filled with closely packed neurofilaments. Clinical and morphological aspects of this new disease entity are discussed with regards to the classical form of giant axonal neuropathy and to other conditions of peripheral neuropathy with giant axons.

Induction of cytochrome P450 isozymes by simultaneous inhalation exposure of hens to n-hexane and methyl iso-butyl ketone (MiBK)

Chickens were exposed simultaneously to the industrial hexacarbon solvents n-hexane and methyl iso-butyl ketone (MiBK). n-Hexane has been shown to be neurotoxic in both humans and other vertebrates. While MiBK is not neurotoxic, it has been shown to greatly synergize the clinical appearance of neurotoxicity in animals exposed to both of these solvents. Groups of hens were exposed for 29 days in inhalation chambers to 1000 ppm n-hexane in combination with 10, 100, 250, 500, or 1000 ppm MiBK. Other groups received either 1000 ppm n-hexane, 1000 ppm MiBK, or ambient air and served as controls. A dose-dependent decrease in body weight and an increase in clinical effects were noted for the highest exposure groups (1000 ppm n-hexane combined with 1000, 500 or 250 ppm MiBK). There was an MiBK dose-dependent increase in cytochrome P450 content and benzphetamine N-demethylase activity, but there was no distinct pattern for ethoxyresorufin O-deethylase or cytochrome c reductase activities. Mixed-function oxidase levels and activities (cytochrome P450 content and benzphetamine N-demethylase) were elevated significantly (P less than 0.05) over controls even in the lowest MiBK group (10 ppm), although there were no clinical signs of neurotoxicity. Four different isozymes of cytochrome P450 were measured immunologically. There was a dose-dependent increase in three of the isozymes, two of which were phenobarbital inducible and one of which was induced by beta-napthoflavone. Quantitatively, the largest increase was in the PB-A isozyme, a phenobarbital-inducible isozyme which accounted for approximately 70% of the cytochrome P450 present in animals treated with MiBK. The results suggest that MiBK selectively induces cytochrome P450 isozymes leading to the metabolic activation of the weak neurotoxicant n-hexane to the potent neurotoxicant 2,5-hexanedione (2,5-HD).

Critical review of the toxicity of methyl n-butyl ketone: risk from occupational exposure

Methyl n-butyl ketone (MBK) was considered rather harmless until an outbreak of peripheral neuropathy occurred in 1973 among workers exposed to MBK. MBK easily penetrates the skin; pulmonary retention is approximately 80-85% in man. Distribution is widespread with highest levels in blood and liver; MBK also reaches the fetal tissues. MBK metabolism probably depends on the route of exposure, and is very similar to that of n-hexane. The critical organ is the nervous system. These effects find expression as peripheral neuropathy, with potential for serious effects of the central nervous system. From the viewpoint of neurotoxicity, 2,5-hexanedione is the most important metabolite. The neurotoxicity is potentiated by several compounds, while MBK itself potentiates the toxicity of other chemicals. From animal experiments, a no-adverse-effect level (NAEL) could not be established. Peripheral neuropathy may develop in workers exposed to only a few ppm of MBK. The difference in the Occupational Exposure Limits for MBK and n-hexane, as established by several organizations, is questioned in view of the neurotoxic effects of these substances.

Measurement precision of a portable instrument to assess vibrotactile perception threshold

The objective of the present study was to evaluate the reliability of a modified version of the commercially available Biothesiometer, and to examine vibrotactile perception thresholds with respect to age and gender. A standardized protocol for measuring vibrotactile perception threshold was administered to 80 subjects, once a week over 4 weeks. Inter-session variability was stable (analysis of variance for repeated measures; P greater than 0.05) and correlations were high (Pearson's: 0.87 less than or equal to r less than or equal to 0.90; P less than or equal to 0.001). For sites on both hands and feet, there was a significant increase with age (0.19 less than or equal to r2 less than or equal to 0.52; P less than or equal to 0.001). Five factor analysis of variance model showed that vibrotactile perception threshold was significantly different with stimulus site, age category and gender; no differences were observed with alcohol consumption or smoking status. The findings indicate that the measurements from this device are highly reproducible and sensitive to expected threshold differences with age and gender. The authors attribute this to technical improvements of the original apparatus, rigid adherence to test protocol and maintenance of standard conditions. This type of instrument would be useful in assessing vibrotactile perception loss in occupational health studies.

Neurotoxicity of organic solvent mixtures: embryotoxicity and fetotoxicity

The toxicity of organic solvents at different concentrations was examined to determine their effect on pregnant rats and their offspring. A very serious effect which inhalational exposure to n-hexane, methyl-ethyl-ketone and a mixture of the two organic solvents produced in pregnant rats was a concentration-dependent increase of intrauterine mortality. In cases of only prenatal exposure, the persistence of reduced body growth was an important result obtained at all concentrations we investigated (first solvent experiment, 500 ppm; second solvent experiment, 800 ppm; and third solvent experiment, 1000 to 1500 ppm; 23 hr day). In these cases, a delay in the maturation of cerebellar cortex was also observed. In cases of pre- and postnatal solvent exposure, the effects of malnutrition were added to the solvent-induced retardation. This resulted in an extreme delay in tissue maturation accompanied by a retarded cell maturation. The solvents examined had no teratogenic effect in pregnant rodents and their offspring. Neurotoxicity was restricted to the known structures of the adult animals, whose axons were damaged. They did, however, show a concentration-dependent embryotoxic and fetotoxic effect.

Reduced performance in tests of memory and visual abstraction in seamen exposed to industrial solvents

Eighty-five seamen exposed to a variety of organic solvents and hydrocarbon compounds during their work on chemical tankers were compared with 59 unexposed seamen. Visual memory, auditory memory and visual abstraction were examined 14 days or more after the last solvent exposure. Multivariate analyses were performed to control the effects of confounding factors such as age, alcohol consumption, smoking, cerebral concussions and educational level. Significant correlations were found between increasing solvent exposure and poor results in tests of auditory memory and visual abstraction. These results indicate adverse psychological effects caused by chronic organic solvent exposure.

Evidence for multiple mechanisms responsible for 2,5-hexanedione-induced neuropathy

The present studies were carried out to investigate the comparative roles of protein cross-linking and alteration in protein phosphorylation in the accumulation of neurofilaments due to aliphatic hexacarbons. In these studies, rats were given 2,5-hexanedione (0, 0.1, 0.25 and 1.0%) for 70 days in their drinking water. In a separate study of in vitro protein phosphorylation rats were given 1% 2,5-hexanedione for 14 days in their drinking water. Spinal cord neurofilaments were isolated and analyzed using sodium dodecyl sulfate-polyacrylamide gel electrophoresis, immunoblotting using anti-neurofilament antibodies, radioimmunoassays (RIAs) of phosphorylated epitopes on neurofilament proteins and protein phosphorylation. Protein cross-linking of neurofilaments was found in all animals treated with 2,5-hexanedione including the lowest dose (0.1%) which did not produce clinical signs of intoxication. Protein phosphorylation of neurofilament proteins, as well as MAP-2 was significantly decreased upon treatment. Protein staining revealed a decreased amount of neurofilament protein and immunoblotting demonstrated neurofilament protein cross-linking in these animals. Protein staining of glial fibrillary acidic protein (GFAP) was unaltered by this treatment. RIAs of phosphorylated and non-phosphorylated epitopes of neurofilament proteins indicated that in vivo phosphorylation of these proteins was also decreased. Two-dimensional gel electrophoresis indicated a shift of the neurofilament proteins to a basic pI, indicating a dephosphorylation of neurofilament proteins. Cross-linked neurofilament proteins also exhibited a pI which was more basic than any of the individual neurofilament proteins. This report demonstrates differential effects of 2,5-hexanedione on neurofilament proteins and indicates that several mechanisms may be responsible for their accumulation.

Abnormalities of the axonal cytoskeleton in giant axonal neuropathy

Intermediate filaments accumulate abnormally in a variety of cell types in individuals with human inherited giant axonal neuropathy (GAN). A characteristic feature of this disorder is the occurrence of focal axonal enlargements filled with accumulations of neurofilaments. The minimum separations between neurofilaments in sural nerve axons of a patient with GAN were 12-30 nm compared with 24-60 nm in controls. The normal sidearm protrusions which cross-bridge adjacent filaments were rare in GAN. Average minimum neurofilament diameter was 12.4 nm in GAN compared with 10.1 nm in controls. Many axons were devoid of neurofilaments and contained an increased density of microtubules, many of which did not run longitudinally. This disorganization of microtubule alignment may reflect the lack of an associated neurofilament lattice. It is concluded that GAN involves abnormalities of neurofilament cross-linkage to one another and to adjacent microtubules. Mechanisms are discussed which could account for this inherited disorder of intermediate filament organization affecting various cell types.

Childhood giant axonal neuropathy. Case report and review of the literature

Giant axonal neuropathy (GAN) is a rare autosomal recessive childhood disorder characterized by a peripheral neuropathy and features of central nervous system involvement. Typically seen are distal axonal swellings filled with 8-10 nm in diameter neurofilaments in central and peripheral axons, and intermediate filament collections in several other cell types. Many neurotoxins produce a morphologically similar neuropathy in humans and experimental animals. Defective nerve fiber energy metabolism has been postulated as a cause in these toxic neuropathies. It is possible that GAN represents an inborn error of metabolism of enzyme-linked sulfhydryl containing proteins, resulting in impaired production of energy necessary for the normal organization of intermediate filaments.

Poisoning by chemical non-medicinal products in Brazil: clinical and laboratory findings in 132 patients

One hundred and thirty-two patients with suspected exogenous intoxication were analysed according to clinical manifestation and toxicological results. Detection of toxic substances was performed in blood, urine and cerebrospinal fluid. The greatest occurrence of chronic exogenous intoxication was in males aged 21 to 50 years old. Peasants and industrial workers were the most affected. 70% of all patients showed neurological manifestations. There was clear predominance of central nervous system impairment associated with poisoning with herbicides and organophosphates.

[Neurologic manifestations in 93 patients with exogenous poisoning caused by non-therapeutic chemical substances]

Ninety-three patients with neurological manifestations and suspected exogenous intoxication were analysed. The detection of toxic substances was performed in the blood, urine and cerebrospinal fluid. There was clear predominance of the central nervous system impairment than the peripheric, and an apparent tropism of herbicides and organophosphates to the central nervous system compared with the peripheral nervous system, and considering the manifestations caused by carbamates and organochlorines. The authors analyse some probable action mechanism of these substances in the nervous system, and show the most frequently neurological signs and symptoms found.

Taxol-induced neuropathy: chronic effects of local injection

The long-term neurotoxic effects of taxol, a compound known to promote microtubule protein polymerization, injected subepineurially into rat sciatic nerve were studied up to 10 weeks post-injection. At the site of injection, taxol caused local axonal reactions and degeneration which were causally related to the slow progressive accumulation of microtubules and other axoplasmic constituents. This culminated in the appearance of giant axonal spheroids and profiles similar to the retraction bulbs of Wallerian degeneration. From these axonal bulbs, many of which arose at nodes of Ranvier, groups of regenerating sprouts emanated. During the acute phase of taxol neurotoxicity, some swollen axons were divested of their myelin sheaths and remained demyelinated for many weeks. After 4 weeks, remyelination was apparent along some fibres. In addition to the accumulation of profiles usually associated with retraction bulbs, there was a vast increase in microtubules, some of which were aligned in concentric rings and formed channels for mitochondria. Microtubule anomalies were also visualized in distal portions of affected fibres and in regenerating sprouts. In contrast, Schwann cells displayed microtubule abnormalities only at the site of the lesion where excessive microtubule polymerization caused the displacement of ribosomes from rough endoplasmic reticulum. Distally, Schwann cells were essentially normal. Axonal depletion and regenerating sprouts were noted further downstream in the tibial nerve, and the gastrocnemius muscle showed changes similar to denervation atrophy. These results extend previous observations by demonstrating chronic, reparatory and reversible phenomena, the implications of which are discussed vis à vis axoplasmic transport and nerve regeneration.

Cross-linking of neurofilament proteins of rat spinal cord in vivo after administration of 2,5-hexanedione

The aliphatic hexacarbons n-hexane, methyl-n-butyl ketone, and 2,5-hexanedione are known to produce a peripheral neuropathy that involves an accumulation of 10-nm neurofilaments above the nodes of Ranvier in the spinal cord and peripheral nerve. In this study, rats were treated with 0.5% 2,5-hexanedione in drinking water for 180 days, and their spinal cord neurofilaments were isolated after development of the neuropathy. Visualization by sodium dodecyl sulfate-polyacrylamide gel electrophoresis revealed a significant reduction in content of the neurofilament triplet proteins in treated animals and the presence of bands migrating at 138K and 260K that were not present in control animals. Analysis of the lanes using immunoblotting procedures and anti-70K, anti-160K, and anti-210K neurofilament antibodies revealed many cross-linked peptides. The 138K band cross-reacted with the anti-160K neurofilament antibody. This suggests that the 138K band is an intramolecular cross-link of the 160K neurofilament subunit. In addition to this peptide, there were numerous high-molecular-weight peptides immunoreactive with all three neurofilament protein antibodies. In addition to cross-linking, there was also a diminished amount of immunoreactive breakdown product of all three neurofilament proteins. This report demonstrates direct evidence of 2,5-hexanedione-induced cross-linking of neurofilament proteins in vivo, which maybe responsible for the accumulation of neurofilament proteins pathognomic of this neuropathy.

Vibratory perception thresholds in shipyard workers exposed to solvents

Vibratory perception thresholds were determined for 90 subjects working at a shipyard with industrial solvents. The duration of exposure varied from 1 to 44 years (mean 19.6 years). Four of the men but none of women had bilaterally significantly increased thresholds in the lower limbs. In addition, seven subjects had increased thresholds in the upper extremities only. Fifteen subjects were further examined at a neurophysiological laboratory to determine the degree of possible peripheral polyneuropathy. The following measurements were made at the laboratory: motor and sensory conduction velocities, electromyography and heart rate variation during normal and forced breathing. The laboratory investigations revealed abnormalities in most of the subjects with increased vibratory perception thresholds in the lower limbs. On the other hand the laboratory measurements were normal for all subjects with increased thresholds in the upper limbs only. It is concluded that when peripheral polyneuropathy is suspected, quantitative vibratory perception threshold measurement is an easily applicable screening method for selecting subjects for further laboratory investigations.

Comparison of central and peripheral nervous system lesions caused by high-dose short-term and low-dose subchronic acrylamide treatment in rats

The effects of high-dose subacute acrylamide treatment of up to 50 mg/kg/day for 4 or 10 d were compared to those of subchronic exposure, up to 12 mg/kg/day for 90 d. In the subacute study, Purkinje cells, long ascending tracts of the spinal cord, optic tract terminal or preterminal regions in superior colliculus, sensory ganglion cells, and distal large-caliber peripheral axons were severely affected. Purkinje cells and fasciculus gracilis changes were the earliest lesions. In the subchronic study, the dominant lesion was confined to the distal peripheral axon, with only minor changes occurring in spinal cord and medulla. Paranodal swellings with the characteristic appearance of neurofilament aggregations were not seen. This morphological study suggests a significant difference between high- and low-dose acrylamide-induced lesions. If there is a reduced tendency for long-term low-dose acrylamide exposure to produce CNS lesions, the risk of irreversible nervous system damage would be less than that predicted from subacute studies.

Analysis of n-hexane, 2-hexanone, 2,5-hexanedione, and related chemicals by capillary gas chromatography and high-performance liquid chromatography

Analytical methods, using capillary gas chromatography and normal-phase high-performance liquid chromatography, were developed for the analysis of the neurotoxic chemicals n-hexane, 2-hexanone, and 2,5-hexanedione and their suspected metabolites. Two gas chromatographic methods, using a 50-m glass capillary OV 101 column and cyclohexane as an internal standard, were employed. In both methods, the injector and detector temperatures were 220 and 280 degrees C, respectively. In method I the following temperature program was used: isothermic at 50 degrees C for 30 min, followed by a temperature increase of 10 degrees C/min to a final temperature of 180 degrees C, which was then maintained for 7 min. This method was used to analyze the following compounds: n-hexane, 2,5-dimethylfuran, 2-hexanone, 3-hexanone, hexanal, 1-hexanol, 2-hexanol, 3-hexanol, 5-hydroxy-2-hexanone, gamma-valerolactone, 2,5-hexanedione, and 2,5-hexanediol. Method II, which was developed for n-hexane and eight of its more common metabolites, used the following temperature program: isothermic at 70 degrees C for 15 min, followed by a temperature increase of 40 degrees C/min to a final temperature of 220 degrees C, which was maintained for 5 min. A linear relationship between peak area and amount injected was observed over a 100-fold range. The minimum detectable amounts ranged from 0.05 to 1 microgram, depending on the compound. Normal-phase HPLC, using a 5-micron silica cartridge fitted into an RCM-100 radial-compression separation system, was utilized to analyze 2-hexanone and its metabolites 2,5-dimethylfuran, gamma-valerolactone, 5-hydroxy-2-hexanone, and 2,5-hexanedione.(ABSTRACT TRUNCATED AT 250 WORDS)

The joint neurotoxic action of inhaled methyl butyl ketone vapor and dermally applied O-ethyl O-4-nitrophenyl phenylphosphonothioate in hens: potentiating effect

The neurotoxic action of inhaled technical grade methyl butyl ketone and dermally applied (O-ethyl O-4-nitrophenyl phenylphosphonothioate (EPN) was studied. Three groups of five hens each were treated 5 days/week for 90 days with a dermal dose of 1.0 mg/kg of EPN (85%) on the unprotected back of the neck. These groups were exposed simultaneously to 10, 50, or 100 ppm of technical methyl butyl ketone (MBK; methyl n-butyl ketone:methyl isobutyl ketone, 7:3) in inhalation chambers. A fourth group was treated only with the dose of EPN and a fifth group with only 100 ppm MBK. The control consisted of a group of five hens treated with a dose of 0.1 ml acetone. Treatment was followed by a 30-day observation period. Simultaneous exposure to EPN and MBK greatly enhanced the neurotoxicity produced when compared to the neurotoxicity produced by either chemical when applied alone. Continued exposure to EPN and MBK resulted in earlier onset and more severe signs of neurotoxicity than exposure to either individual compound. The severity and characteristics of histopathologic lesions in hens given the same daily dermal dose of EPN in combination with inhaled MBK depended on the MBK concentration. Histopathologic changes were more severe and prevalent in the 100 ppm MBK:1 mg/kg EPN group than in the others. In this group, Wallerian-type degeneration was seen along with paranodal axonal swellings. The morphology and distribution of these lesions were characteristic of those induced by MBK. In the 50 ppm MBK:1 mg/kg EPN group axonal swelling was evident but not clearly identifiable as paranodal. Hens treated with 10 ppm MBK:1 mg/kg EPN had minimal lesions with low incidence of axonal swellings. These were not as large as those seen in MBK neurotoxicity, but instead resembled the histopathologic lesions caused by EPN. The results indicate that the combined treatment gave a value for neurotoxicity coefficient which was two times the additive neurotoxic effect of each treatment alone. Pretreatment with three daily ip doses of 5 mmol/kg technical grade MBK or methyl n-butyl ketone (MnBK), equally increased chicken hepatic microsomal cytochrome P-450 content. Also, hepatic microsomes from MBK-treated hens metabolized [14C]EPN in vitro to [14C]EPN oxon to a much greater extent than those from control hens. These results suggest that MBK potentiates the neurotoxic effect of EPN, at least in part, by increasing the metabolic activation of EPN to the more neurotoxic metabolite EPN oxon.(ABSTRACT TRUNCATED AT 400 WORDS)

Hereditary motor sensory neuropathy type II with neurofilament accumulation: new finding or new disorder?

Peroneal muscular atrophy is now known to be heterogeneous and to be due to various underlying genetic mechanisms. Exploring this heterogeneity further, we report on a German kinship with the clinical, genetic, and nerve conduction features of hereditary motor and sensory neuropathy type II (HMSN type II) but whose sural nerves on biopsy were found to show infrequent axonal swellings with neurofilament accumulations not previously described. The dominant inheritance and absence of kinky hair set this disorder apart from giant axonal neuropathy. There was no history of toxic exposure to industrial chemicals. We conclude that the disorder either is a new type of HMSN or is HMSN type II with previously unencountered neurofilament accumulations. Neurofilament accumulation indicates that the axon could be a site for primary derangement and may implicate an abnormality of slow axonal flow. In addition, some of the patients exhibited features suggestive of a cardiomyopathy.

Neurophysiological studies on the relation between the structural properties and neurotoxicity of aliphatic hydrocarbon compounds in rats

In order to determine the specific structural properties responsible for neurotoxic activity, the comparative neurotoxicity of n-hexane, methyl n-butyl ketone, 2,5-hexanedione, and their relatives was investigated in the peripheral nerves of rats. The maximum conduction velocity of motor and sensory fibres and the motor distal latency of the tail nerves of rats were periodically examined in animals receiving repeated subcutaneous injections of 11 aliphatic monoketone or diketone compounds and their relatives for prolonged periods. A study of the comparative neurotoxicity of n-hexane, methyl n-butyl ketone, and their metabolites showed that 2,5-hexanedione was the most actively neurotoxic. Furthermore, a study of other symmetrical diketones with different carbon numbers showed that 2,4-pentanedione, which is structurally similar to 2,5-hexanedione, possessed a different type of neurotoxic activity than 2,5-hexanedione. Regarding aliphatic monoketone compounds, acetone, 2-pentanone, 2-heptanone, and 2-octanone were confirmed non-neurotoxic for the peripheral nervous system. Evidence from some previous reports, however, suggested that 3-heptanone, 4-octanone, and 5-nonanone might produce neuropathies by being converted to 2,5-diketones under specific conditions.

Ultrastructure of carbon disulphie neuropathy

Adult Wistar rats were exposed to carbon disulphide (CS2) vapour at a concentration of 2.4 mg/l of air for 5 days a week (6 h a day), and the ultrastructure of peripheral nerves, neuromuscular junctions and muscles was investigated after 6 months of exposure to CS2. Numerous giant axons, i.e. paranodal or internodal swellings, were seen in the peripheral nerves. At the swollen paranodes, the myelin sheath was thinned, in other regions large intramyelinic vacuoles indicative of more dramatic demyelination were observed at axonal enlargements. Axonal enlargements consisted essentially of whorls of tightly packed neurofilaments. A number of nerve fibres underwent complete degeneration, but at the same time there was evidence of nerve regeneration. Nerve terminals were affected in a similar way following CS2 exposure. At neuromuscular junctions, filamentous swellings of nerve terminals preceded their degeneration and eventual denudation of synaptic gutters. As a rule, the postsynaptic part of neuromuscular junctions remained unimpaired by CS2 treatment. Muscles were affected by both atrophy and degeneration. Clusters of dense and lamellar bodies and numerous autophagosomes indicative of direct myotoxic effect of CS2 were frequently encountered in the investigated muscles. Some muscle fibres apparently underwent necrosis judging from the occurrence of myotubes characteristic of muscle degeneration and regeneration. The pathomorphology of CS2 neuropathy resembles that of other toxic neuropathies which presumably have a common origin in impaired energy metabolism.

Cerebral metabolic and circulatory effects of 1,1,1-trichloroethane, a neurotoxic industrial solvent. 1. Effects on local cerebral glucose consumption and blood flow during acute exposure

The effects of inhaled 1,1,1-trichloroethane (3500, 6000, and 7800 ppm) on behavior, local cerebral blood flow, and local cerebral glucose consumption were studied in awake rats. The effect of the solvent inhalation on the EEG pattern and local cerebral blood flow was also studied in paralyzed animals under N2O analgesia. Exposure of awake animals to 6000 ppm 1,1,1-trichloroethane induced a decrease in motility and exploratory behavior. At 7800 ppm the rats were clearly ataxic. The local cerebral glucose consumption in 23 brain regions was studied by the [14C]deoxyglucose technique. A decrease was observed ranging from 14 to 55% of control values. The inferior colliculus and substantia nigra displayed the largest reductions. In exposed animals the local cerebral blood flow increased in 11 brain structures by 28-45%. In animals under N2O analgesia, 7400 ppm 1,1,1-trichloroethane induced a depression of the EEG activity. In these animals the local cerebral blood flow increased by 12-99%, with a large variability in blood flow between the different structures. It is concluded that exposure of rats to subanesthetic doses of 1,1,1-trichloroethane induces an increase in cerebral blood flow in spite of a concomitant decrease in glucose consumption and depression of cerebral function.

Further studies on ketone neurotoxicity and interactions

Ethyl n-butyl ketone (EBK, 3-heptanone) has not been reported to produce neurotoxicity in previous studies when it was given by oral or inhalation routes. In the present study, EBK given by gavage at 2 g/kg/day, 5 days/week for 14 weeks produced a typical central-peripheral distal axonopathy characterized by "giant" axonal swelling and neurofilamentous hyperplasia. This dose approximates the single dose LD50 (2.76 g/kg) of EBK determined by Smyth et al., (J. Ind. Hyg. Toxicol. 31, 60-62, 1949). Large multiple doses (1.5 g/kg/day, 5 days/week, 14 weeks) of methyl ethyl ketone (MEK) given by gavage potentiated EBK neurotoxicity but 5-methyl-2-octanone did not. MEK modestly increased the urinary excretion of two neurotoxic gamma-diketones, 2,5-heptanedione, and 2,5-hexanedione, when MEK was given by gavage with EBK. When rats were exposed to EBK (700 ppm) and MEK (700 or 1400 ppm) in combination by inhalation exposure, serum 2,5-heptanedione levels were increased approximately 2.5 times. This effect was absent at MEK levels of 70 ppm. The serum from rats exposed to EBK or EBK/MEK combinations did not contain 2,5-hexanedione. The toxicity of EBK appears to involve the metabolism of EBK to two neurotoxic gamma-diketones, 2,5-heptanedione and 2,5-hexanedione. Combined EBK/MEK exposure modestly increased gamma-diketone levels in the serum and urine suggesting that MEK potentiates EBK neurotoxicity by stimulating the metabolism of EBK to neurotoxic metabolites. The magnitude of the doses used in the present study to produce neurotoxicity and the absence of neurotoxicity in previous subchronic studies suggest that the neurotoxic hazard of EBK is low.

Toxic polyneuropathy of shoe workers in Italy. A clinical, neurophysiological and follow-up study

Cases of polyneuropathy due to exposure to industrial solvents have been studied at several shoe factories in the province of Siena. After the screening of 654 employees 98 verified cases were detected. Of these, 16 were rated as moderate to severe, 45 as mild, and 37 were minimally involved but with characteristic electrodiagnostic abnormalities. Follow-up study in 53 patients showed that neurological signs and symptoms as well as electrodiagnostic abnormalities continued for years in several patients. In addition, after a year's observation, some patients showed signs of central nervous system dysfunction such as spasticity of the lower limbs and increased deep tendon reflexes. High percentages of commercial n-hexane were found in all the samples of glues and solvents collected from home-workers and from factories where cases of polyneuropathy occurred.

Distortions of the nodes of Ranvier from axonal distension by filamentous masses in hexacarbon intoxication

A study has been made of the structural changes of nodal and paranodal regions of the nodes of Ranvier of peripheral nerves of rats in which marked accumulations of neurofilaments have occurred within axons under the influence of 2,5-hexanediol over 10 weeks. The neurofilamentous masses caused distension of the axon at two points of apparent weakness as they attempted to slide through the axonal constriction at the nodes. Principally, a spiral axonal protrusion pushed into the zone of unattached myelin loops in the proximal paranodal spinous bracelet of Nageotte. This led to a conical widening of the paranodal constriction and considerable attenuation of the overlying myelin. No degeneration of the myelin occurred however. Alternatively, or additionally, a protrusion occurred of the axon at the nodal region which increased the nodal gap width and occasionally compressed and displaced the adjacent distal paranodal constriction which could have led to some obstruction of axoplasmic flow. Swelling of distal paranodal regions occurred later and was usually associated with proximal swelling. It was also accompanied by evidence suggesting transnodal passage of filamentous material. Sometimes, however, striking nodal constriction occurred in association with symmetrical paranodal swelling. These observations suggest that the spiral glial-axonal relationships at nodes of Ranvier are capable of marked deformation that might allow the intra-axonal neurofilamentous masses to move distally. These findings are discussed in relation to the structural features of the paranodal constrictions.

Morphometric evaluation of degenerative and regenerative changes in isoniazid-induced neuropathy

Morphometric studies of the pathologic changes were carried out on the peripheral nerves, spinal roots, and different levels of the Goll's tract in rats given isoniazid and killed 1, 2, 3, 4, 5, 6, 7, 14, and 30 days after intoxication. In teased fiber preparations, axonal degeneration was the main change present, and this was seen as early as day 2 in the peroneal and distal sural nerves. The frequency of myelinated fibers showing axonal degeneration was higher in the distal than the proximal sural nerve, and in the ventral than the dorsal root. In the group of rats killed on 5, 6, 7, and 14 days, a significant decrease of the myelinated fiber density was observed in the distal and proximal sural nerves, ventral root, and at the third cervical level of the Goll's tract. The degree of fiber degeneration was more severe in the distal than in the proximal sural nerve and in the third cervical than the fifth thoracic level of the Goll's tract. Preferential decrease of large myelinated fibers was noted in all the affected nerves. No definite abnormalities, however, were seen in nerve cells of the 6th lumbar spinal ganglia and anterior horn cells of the lumbar spinal cord on light microscopy. On 30 days, regeneration at varying degrees was discerned in all the affected nerves with significant increase of small myelinated fibers, particularly in the ventral root. The findings indicate that both centrally and peripherally directed myelinated axons are more affected in the distal than in the proximal segments while the neuronal cell bodies are spared. The spatio-temporal evolution of this pattern of change is compatible with the concept of the "dying back" process or central-peripheral distal axonopathy.

The early evolution of neurofilamentous accumulations due to 2,5-hexanediol in the optic pathways of the rat

Rats were given 2,5-hexanediol in their water for more than 5 weeks. The changes in the optic pathways were studied both qualitatively and quantitatively. Increase in 10-nm filaments within axons was noticeable from 10 days onwards in the superior colliculus and in the brachium of the superior colliculus. From then onwards there was a steady increase in the number of affected axons, and their gross enlargement occurred. The proportion of affected fibres in the brachia was in fact few, but all fibres were of retinotectal origin. Less than half of those in the superior colliculus that were swollen were of retinal origin. Measurement of axon diameter versus myelin sheath thickness showed gross relative thinning of the latter. Axon degeneration did not occur but there was increasing ultrastructural evidence of impairment of transport of organelles both centrifugally and centripetally as the filamentous masses accumulated.