Keywords neurocysticercosis - hydrocephalus - neurons - experimental model
Palavras-chave neurocisticercose - hidrocefalia - neurônios - modelo experimental
Introduction
Neurocysticercosis (NC) is a common parasitic infection of the central nervous system
(CNS) prevalent in Latin America, Sub-Saharan Africa, and Southeast Asia. Despite
having a possibility of eradication, the disease remains endemic in communities with
poor sanitary conditions and has reemerged in developed countries because of migratory
flows.[1 ]
[2 ]
[3 ]
The course of the disease is dependent upon the number of parasites, their stage of
development, distribution within the brain, the subarachnoid space and the ventricles,
and parasite-host interaction.[4 ] In the extraparenchymal form of the disease, the cysts of the tapeworm Taenia solium lodge in the cerebrospinal fluid (CSF) compartments and may cause meningitis, vasculitis,
hydrocephalus, and raised intracranial pressure.[5 ]
[6 ] Hydrocephalus is one of the most devastating complications related to extraparenchymal
NC. The high rate of infections and malfunctions of the ventricular shunts used to
release hydrocephalus in patients with NC are the leading causes of the considerably
high mortality in these patients.[7 ] Neurocysticercosis-associated hydrocephalus may occur due to obstruction caused
by the presence of cysts at the narrow point of the CSF flow pathway within the brain
ventricles, or due to inflammation within the subarachnoid cisterns.[8 ]
[9 ]
[10 ]
The pathophysiology of NC-induced hydrocephalus and the therapeutic options for the
treatment of this condition are well-studied. However, the heterogeneity of the disease
among different populations and age groups, mainly concerning the clinical presentation
and the response to medical treatment, is poorly understood.[11 ] Experimental animal models of NC-induced hydrocephalus might help to address some
of these issues.
Recently, we have developed a rat model of NC-induced hydrocephalus by cisternal inoculation
of Taenia crassiceps cysts. In this model, we observed obstructive and inflammatory features that confirmed
the successful induction of hydrocephalus.[12 ] However, the role of inflammation in NC-induced hydrocephalus is not yet studied.
Since in clinical practice, it is common to find patients with NC-related hydrocephalus
without viable cysts, it might be possible that the degenerating cysts induce hydrocephalus.
The present study aimed to assess whether cisternal inoculation of T. crassiceps cyst antigens in different concentrations could induce hydrocephalus and tissue damage
of different severities.
Methods
Experimental Animals
Sixty male Wistar rats (Rattus norvegicus) aged 6 weeks were used in the present study. The animals were handled according to
the current guidelines for the care and use of laboratory animals, and the local institutional
review board approved the study project. The animals were kept under adequate sanitary
conditions with food and water available ad
libitum , in a room with a 12 h light/dark cycle under controlled temperature (21°C).
The animals were randomly divided into 2 groups: low (n = 30) and high concentration
(n = 30) T. crassiceps antigens inoculation groups. The animals underwent magnetic resonance imaging (MRI)
at 1, 3, and 6 months after the inoculation. After the 3rd MRI, the animals were euthanized and the brains were collected for histological assessment.
T. crassiceps Antigens Preparation
T. crassiceps cysts were maintained by subsequent inoculations in the peritoneal cavity of mice,
where the cysts reproduce by gemmulation.[13 ] The cysts were aseptically removed from the peritoneal cavity of mice and collected
in beakers for immediate preparation of the antigenic suspensions. For the low concentration
group, the cysts were resuspended in 100 ml of saline and for the high concentration
group no diluent solution was added. The cysts suspension was sonicated with 60% amplitude
at 10°C until a homogenous suspension was obtained. The resulting suspensions were
centrifuged at 0°C and 180 Hz for 5 minutes and the supernatant was collected. The
concentration of the protein in the suspensions was determined by the Bradford method.
The protein concentration in the low concentration group was 2.4 mg/ml, and in the
high concentration group, it was 11.6 mg/ml.
Inoculations
The rats were anesthetized with an intraperitoneal injection of 0.1 ml/kg mixture
of ketamine (100 mg/ml) and xylazine (20 mg/ml). A 1-cm long skin incision was made
at the occipitocervical junction, followed by blunt dissection pf planes until reaching
the skull, the posterior arc of the first vertebra, and the atlanto-occipital membrane.
Next, 0.02 ml of the cyst suspension was injected with a 25-G needle between the occipital
bone and the first vertebrae, through the atlanto-occipital membrane towards the cisterna
magna. The skin was sutured with a 4.0 mononylon suture.
Magnetic resonance imaging
Magnetic resonance imaging was performed using a Vet-MR 0.25 T equipment (Esaote,
Santo André, SP, Brazil). For MRI examination, the animals were anesthetized with
0.5 ml/kg of the aforementioned mixture of xylazine and ketamine. The image acquisition
protocol consisted of T2-weighted echo gradient acquisitions (slice thickness: 0.6 mm,
echo time: 5 ms, repetition time: 10 ms) which provided a good visualization of the
dilated ventricles, as previously described.[12 ] Ventricle volume was determined using ITK-SNAP version 3.0.0 software (GNU General
Public License, Philadelphia, PA, USA) with a manual segmentation of the ventricles
frame-by-frame. The investigator was blinded to the animal groups.
Morphological Assessment
After the 3rd MRI (6 months postinoculation), the animals were euthanized with an overdose of xylazine
and ketamine. The animals were transcardially perfused with saline to wash out blood
vessel contents, followed by 10% buffered formalin for tissue fixation. The brains
were dissected out of the skull and cut in the coronal plane at the level of the optic
chiasm. The samples were fixed in formalin overnight and then were dehydrated by transferring
the tissue through solutions of increasing alcohol concentrations. Following this,
the tissues were diafanized in xylene and embedded in paraffin. The blocks were sectioned
into 5 µm thick sections and the sections were stained with hematoxylin-eosin. The
morphological assessment was carried out using a morphometric approach using stereological
point-counting, by which a uniform grid is plotted over the image and the points crossing
the lines are counted.[14 ] The morphometric approach considered the mean number of neurons, oligodendrocytes,
astrocytes, vessels, inflammatory cells, and lymphocytes in the periventricular and
ependymal regions.
Statistical Analysis
The distribution of data was assessed with the Shapiro-Wilk test. Differences between
the groups were assessed using the Student t-test or the Mann-Whitney test. Differences
in repeated measures within the same groups were assessed using the paired t-test
and the Wilcoxon test. The Fisher exact test was used to compare frequencies. Differences
were considered significant for p -value < 0.05. All statistical analyses were performed using IBM SPSS Statistics for
Windows, version 21.0 (IBM Corp, Armonk, NY, USA).
Results
The periprocedural mortality was 9 and 19 in the low and high concentration groups,
respectively. Among the remaining 21 rats in the low concentration group, 4 died during
the follow-up and 1 after the 3rd MRI. In the high concentration group, from the 11 remaining animals, 3 died during
the follow-up period. Thus, imaging and histological assessment were performed on
the remaining 16 rats in the low concentration group and 8 in the high concentration
group.
Slight enlargement of the brain ventricles was observed in both groups postinoculation;
the increase was more evident and significant in the high concentration group ([Fig 1 ]), mainly at 6 months postinoculation. In the low concentration group, a significant
increase was observed between 1 and 3 months postinoculation (medians 1.78 mm3 and 2.15 mm3 , respectively), while in the high concentration group, a significant increase was
observed between 3 and 6 months postinoculation (medians 1.89 mm3 and 3.86 mm3 , respectively) ([Table 1 ]). In the 1 and 3 months postinoculation follow-ups, the volumes of the ventricles
were not different between the 2 groups (p = 0.180 and 0.397, respectively). However, a significant difference in the ventricle
volume was observed between the 2 groups at 6 months postinoculation (medians 2.00
versus 3.86 mm3 ; p = 0.003).
Table 1
Ventricle volumes (mm3 ) in the low- and high-concentration groups at 1, 3, and 6 months postinoculation
Month
Group
Average
Standard deviation
Median
Range
p-value
1
Low
1.67
0.85
1.78[† ]
0.30–2.82
0.180[* ]
High
2.64
2.25
2.04[‡ ]
1.04–8.08
3
Low
2.62
2.79
2.15[† ]
0.22–12.83
0.397[* ]
High
3.16
3.58
1.89[‡ ]
0.45–11.20
6
Low
2.63
2.52
2.00[† ]
0.37–11.57
0.003[* ]
High
5.26
3.35
3.86[‡ ]
2.37–12.68
† Wilcoxon test for repeated measures. In the low-concentration group, ventricle volumes
were different at 1 and 3 months (p = 0.032), but not at 3 and 6 months (p = 0.984).
‡ Wilcoxon test for repeated measures. In the high-concentration group, ventricle volumes
were not different at 1 and 3 months (p = 0.646), but were different at 3 and 6 months (p < 0.05).
* Mann-Whitney test for independent samples. The ventricle volumes were not different
between the two groups at 1 and 3 months, but were different at 6 months.
Fig. 1 Coronal magnetic resonance images (a) in the low-concentration group the ventricles
cannot be visualized; (b) slight enlargement of the ventricles (dashed circles) in
the high-concentration group; (c) an evident ventricle enlargement (arrow) from the
high-concentration group.
Morphological assessment revealed a few lymphocytes and new vessels in both groups.
The ependymal lining was normal in both groups. The mean oligodendrocytes density
was 9.72 and 5.18 in the low and high concentration groups, respectively (p < 0.001). Similarly, within the periventricular area, a significant difference in
the neuronal density was observed between the 2 groups (21.00 versus 15.69, low versus
high concentration group, p < 0.001, [Figure 2 ]). Conversely, there was no significant difference in astrocyte density between the
two groups (2.89 versus 2.74, low versus high concentration group, p = 0.73).
Fig. 2 Hematoxylin and Eosin staining of the brain sections depicting differences in neuronal
density (a) the low-concentration group has a higher neuronal density compared to
(b) the high-concentration group.
Discussion
Neurocysticercosis is a preventable, but an embarrassingly neglected infectious disease
still prevalent in nondeveloped countries. It remains endemic in several countries
in Latin America, sub-Saharan Africa, and Southeast Asia.[15 ] Experimental animal models of neurocysticercosis are a valuable tool to study the
characteristics of inflammation and the basic mechanisms underlying the heterogeneous
relationship between the parasite and the host.[16 ]
In a previous study, we observed that, in rats, cisternal inoculation of antigenic
suspension of T. crassiceps cysts induced a slight increase in the ventricle volume, without any signs of inflammation.
We speculated that it was due to the low concentration of the inoculated proteins
and that the inoculation of a higher concentration of antigens could induce a more
intense inflammatory response, and more evident hydrocephalus.[17 ] Herein, we observed that the animals who received a higher concentration of T. crassiceps cyst proteins had a significantly higher ventricle volume than those who received
a low concentration of cyst proteins at 6 months postinoculation, confirming the previous
hypothesis. In our experimental model, the induction of hydrocephalus at 6 month postinoculation
is dependent on the presence of living cysts.[17 ] The live T. crassiceps cysts might continuously recruit the inflammatory cells and obstruct the narrow passages
within the ventricles; thus, a combination of inflammation and mechanical obstruction
may lead to hydrocephalus.[17 ] In contrast, the antigenic suspension form of T. crassiceps cysts may evoke a transient inflammatory response that may cause slight enlargement
of the lateral ventricles without obvious hydrocephalus. It might be possible that
inoculating the animals with even a higher concentration of cyst proteins may lead
to hydrocephalus; however, further studies are required to prove this other hypothesis.
Another possible explanation is that this slight ventricle enlargement is due to a
brain atrophy caused by the inflammation evoked by the response against the parasite.
Actually, the intraperitoneal inoculation of T. crassiceps is able to impair the learning performance of mice, which is accompanied by hippocampal
sclerosis. Apoptosis of hippocampal cells may be related to a breakdown of the blood-brain
barrier determined by the presence of circulating T. crassiceps metacestode factor.[18 ]
Clinically, it has been well-documented that dead cysts may cause arachnoiditis, ependymitis,
and hydrocephalus,[19 ] which is one of the reasons for using corticosteroids to control inflammation during
the antiparasitic treatment.[20 ] A few studies have cautioned about the risk associated with intraoperative rupture
of T. solium cysts, which may lead to ventriculitis[21 ]
[22 ]; however, recent studies downplayed these risks.[23 ]
[24 ]
[25 ] In the surgical management of hydatid cysts, the rupture of the cyst may worsen
inflammation and lead to the seeding of infection.[26 ]
[27 ]
In the present study, inoculation of the animals with a higher concentration of T. crassiceps cyst antigens failed to induce remarkable ependymitis and hydrocephalus. A possible
explanation is that inoculation does not mimic the natural degeneration of cysts observed
in humans. Further studies on cyst degeneration using antiparasitic drugs are required
to conclusively prove this. Another possibility is the obvious differences between
the parasites (T. solium versus T. crassiceps ) and the hosts (human versus rat).
We observed no significant difference in the number of inflammatory cells between
the two groups. Moreover, the number of lymphocytes was low in both groups. Since
the histopathological assessment was performed at 6 months after the inoculation,
the possibility of a transitory and self-limited inflammation that followed the inoculation
within the first few days cannot be ruled out.
Interestingly, we observed reduced oligodendrocytes and neuronal density in the periventricular
area of animals from the high-concentration group, which may explain the neuropsychological
changes observed in patients with neurocysticercosis. The association between dementia
and neurocysticercosis was demonstrated a few years ago; patients with neurocysticercosis
may present changes in working and episodic verbal memory, executive functions, naming,
verbal fluency, constructive praxis, and visuospatial orientation.[28 ]
[29 ]
[30 ] Recently, another group with interest in experimental models of neurocysticercosis
showed increased demyelination and hippocampal disorganization following intraventricular
inoculation of T. crassiceps cysts.[31 ] However, these animals also presented ventricle enlargement, which might have caused
the observed morphological changes.[32 ]
[33 ] In the present study, despite the low volume of the ventricles, the animals with
a higher concentration of cyst antigens presented a reduced neuronal density. Further
studies are required to investigate the link between neuroinflammation and neurodegeneration,
as well as between neurocysticercosis and mesial temporal sclerosis, which has been
extensively demonstrated in clinical studies.[34 ]
[35 ]
[36 ] Besides, a recent clinical study has shown an interesting correlation between anti-brain
protein autoantibodies and the levels of secreted T. solium glycoprotein HP-10, suggesting that the level of stimulation of the autoantibody
response may be a function of the number of viable parasites.[37 ] This information adds evidence to the link between neurocysticercosis and neuroinflammation.
Finally, the present study has limitations. First of all, we did not use a control
group without interventions. However, in our experience with previous experiments,
we found that the inoculation of saline leads to no change on MRI and histologic assessments.
Second, we could not precise the extent of inflammation soon after the inoculations
because histologic assessments were done only at the end of the experiments. It is
possible that the animals with larger ventricles and more histologic impairments had
a more exacerbate initial inflammatory response, but we cannot be certain of this
point.
In conclusion, we found that inoculation of a higher concentration of T. crassiceps antigens in the subarachnoid space of rats leads to a more remarkable ventricle enlargement
and a significant reduction in the neuronal density within the periventricular area.