Immunohistochemistry technique

First, antigen retrieval was carried out by heating the slides in a microwave oven for 20 minutes at 750 W in Tris-EDTA buffer (Tris 10mM, EDTA 1mM, pH9). After washing in PBS (Buffered saline phosphate, 100 mM phosphate pH 7,2; 150 mM NaCl), sections were incubated consecutively in: 1) primary antibody (1h) β-APP (Invitrogen, CA, USA, 1:50); 2) Peroxidase blocker (Dako) 10 minutes, 3) EnVision dual rabbit, mouse (Dako) 30 min, 4) diaminobenzidine. Liquid DAB+Substrate Chromogen System (Dako) 10 minutes as commercial instructions recommend (50 mL DAB in 1mL of substrate buffer). After each step, sections were washed twice during 5 minutes with PBS, and after DAB, they were washed with distilled water.

Solutions of primary antibody β-APP were prepared with a commercial antibody diluent (Dako). In every case, negative controls were included, in which primary antibody was replaced with PBS. At the end sections were counterstained with Harris Hematoxylin for 30 seconds.

Distributions patterns of the expression of -app

First is important to know that amyloid pathology is different in each case of TBI because it is spatially heterogeneous depending on the sort, intensity, and location of the impact.

A simple but very practical classification of TBI is that one which divides it in focal or diffuse.

Focal TBI involves injuries that affects the cortex, which means head trauma with intracranial hemorrhages and/or rise of intracranial pressure.

Diffuse axonal injury involves (DAI) is consequence of rotational forces secondary to acceleration and deceleration which causes axonal disruption in the white matter. (²⁶)

All this suggests that the kind of insult plays a very important role in the process of β-APP accumulation. (²⁷).

In fact, experimental models that cause DAI using rotational forces at different and controlled radians per second, observed different focal and chronological expression of amyloid proteins in each group (²⁸). Likewise, there are significant differences in the results of studies that use weight-drop model and those that use fluid percussion model (²²).

On the other hand, hypoxia per se is not a cause of axonal damage but it is a consequence of vascular complications due to the rise of intracranial pressure and brain herniation. Given that β-APP transport is an active process (it consumes energy), it seems reasonable to assume that it accumulates also in cases of metabolic issues, as hypoglycemia, global hypoxia, infections and brain swelling. (²⁹) This is why the assessment of β-APP allows for the diagnosis of the increase of intracranial pressure even when it is not high enough to cause herniation. The origin of this situation (cardiac arrest, CO poisoning, metabolic issues...) can be determined by the clinical context. (³⁰)

Given that the accumulation of this protein is not a consequence of a unique injury mechanism, some authors (3,²⁹) have searched for features that are unique to each type of injury. The pattern of distribution of β-APP can be useful to avoid false positives of traumatic causes of death. These authors have described three patterns:

It is mostly assumed that diffuse multifocal pattern of axonal varicosities is related with a traumatic cause of brain injury; the Z pattern is related with hypoxia and mixed pattern is associated with severe trauma that gives place to brain swelling or respiratory distress that adds a hypoxic damage to trauma.

This is consistent with our results: we observed how change the pattern of expression (especially in the gray matter) as the time of survival is longer, since inflammation and brain swelling take a minimum time to begin.

Onyszchuk et al. (²³) settled a scale of intensity of staining related with the severity of trauma in cases with a survival period not longer than 2 h. It ranged since no staining at all, until mass staining of axonal bulbs. In grades 0-2 the immunohistochemistry could be very weak and may be due to ischemia. It is the time elapsed till death who determine the evolution of DAI and consequently the staining grade. When the survival period is shorter than 35 minutes, the intensity of staining is not more than a grade 2.

Our findings are not about the extension of the secondary brain damage, because we do not work in experimental conditions, but in daily forensic work conditions. Therefore we only sample the region of direct impact and thus we are not able to diagnose the extension of hypoxia which is probably that Onyszchuk observed. This point is relevant for clinicians, but it is not for postmortem diagnosis, so, given that the earliest changes occur in the gray matter, the study of this region is useful to relate the cause of death with mild head trauma where there is not an obvious macroscopic lesion. All this means that the presence of β-APP allows solving diagnosis questions when it is considered under its clinical and forensic context, as other authors also settled (³¹,³²).

In fact, we were able to diagnose which head trauma was responsible of the death in one case of our series. At first, this case was included in the group with a survival period longer than 24 hours (five days), since the patient fell accidentally and on went a head trauma with neurological symptoms (unconsciousness), but she declined medical assistance. Five days later, she fell again and suffered another head trauma, but this time there was nothing wrong with the neurological assessment (Glasgow scale assessment 15). Shortly after meanwhile she was on going a CT, the Glasgow scale descended to 3 and she died some minutes later.

The CT showed an occipital skull fracture and a fatal intracranial hemorrhage. The first head trauma was clinically more serious, so the diagnose was a subacute subdural haematoma and therefore the first injury was considered the cause of death. Nevertheless, when we studied the brain with -APP, the pattern of expression was consistent with an evolution no longer than 2 hours. There was not any secondary axonal disruption that may suggest any rise of ICP and secondary hypoxia. All this pointed to an acute hemorrhage and death was a result of the second trauma. In this case there was not any legal consequences, but this assessment could be crucial in murders or accidents.

On the other hand, it is said that it is necessary an all brain wide sampling to diagnose DAI and see concomitant morphological changes that may suggest ischemia as eosinophilia and neuron necrosis in the region C1 of the hippocampus, which is the most sensitive region of the brain to hypoxia (³³). Other authors conclude that it is necessary, at least, obtain 6 samples of the cortex of both hemispheres to diagnose DAI. (³⁴)

Nevertheless, the study of samples from the contusion area is useful to determine the severity of trauma and the survival period, and it does not require special training or equipment, what allows gather them even in adverse conditions. In addition, the assessment of the contusion area has a high confidence level in forensic diagnosis, because the changes in this zone are more obvious and significant since this point is the ''ground zero'' from where the damage becomes widespread. (³⁵)

Chronology of the β-app expression after trauma

It must be considered that DAI and protein accumulation are processes that take place in two phases: in the acute one, the insult provokes axonal disruption, especially in the cortex. This triggers protein accumulation. In the chronic phase, vascular changes difficult the clearance of these proteins, thus close this circular process of DAI which gets spread in regions not directly affected by injury. (²⁷)

It has been established that there is a relationship between the immunoreaction for β-APP and the time elapsed since trauma to death. There is not any significant reaction to this marker in cases where survival time is shorter than 30 minutes, whatever the cause of death may be. In addition, the penumbra period in which β-APP is not detectable, gets shorter as the severity of trauma increases. (²⁹)

In general, it is accepted that, in patients elder than 1 year, DAI cannot be diagnosed before than 1,5 hours post-injury. (¹⁷)

In earlier forensic studies, axonal bulbs are not observed in cases where survival time is shorter than 6 hours. This is the point in time in which β-APP and neurofilaments accumulate in the soma of neuron. (³⁶). Far from the area directly affected by trauma, swelling axons and varicosities (stained with β-APP and neurofilaments markers) are present since the 3d to 7th days after injury. They begin to disappear after 6 months of the trauma. Nevertheless β-APP is positive 3 hours after injury in the contused area and it is still positive 6 months later in parenchyma and around vessels (³⁷). Other authors proved the presence of β-APP staining 24 hours after trauma in swelling neurons that undergo dystrophy of their fibers. (¹⁵)

In cases of mild trauma, the β-APP shows a specific chronological behavior: its first immunoreactivity takes place between 1.5 and 3 hours after trauma. It reaches a maximal peak of expression at 10-15 hours, and it stays stable during days or weeks. (¹⁴)

Chronological differences between our results and those of other authors may be due to that they observe the expression of this marker in brain regions far from that directly affected. It means that they study the damage secondary to hypoxia and this phenomenon is time dependent according to the severity of trauma. Moreover, the sensibility to hypoxia in each part of the brain is different, so that they also show chronological differences expressing β-APP

Kelley (¹⁸) studied how axonal transport gets disturbed through the assessment of the β-APP expression in the thalamic nucleus after a controlled axotomy. This encephalic structure is very sensitive to hypoxia and in consequence to secondary axonal damage: fifteen minutes after axotomy the thalamic nucleus shows immunoreactivity to this protein. The diameter of the axons appears a little enlarged and since 60 minutes after trauma axonal bulbs begin to appear.

The first regions where β-APP is seen are the parasagittal tract and structures of the mid line of the brain including the caudal zone of the corpus callosum (³⁸). Thus, the most common distribution of β-APP immunoreactivity involves the brain stem (95% of cases), the internal capsule and the thalamus (80% of cases), the corpus callosum and the parasagittal white matter (70% of cases). As we can see more expression of β-APP other than hemorrhages, it can be said that its accumulation does not depend on the presence of hemorrhages. (¹⁴). Our results confirm all this, since this marker was positive in all regions which showed external signs of contusion, regardless the severity of trauma.

In the pons, it is not possible to see any reaction for β-APP before three hours after trauma, regardless of the presence of subdural hematoma or cortical hemorrhage, which would demonstrate the vitality of the lesion. β-APP accumulates even in cases of non-mechanical trauma, but in cases of hypoxia or ischemia the expression of β-APP takes place later (8 to 24 hours after the insult) and the staining is weaker than in cases of mechanical trauma.

Diffuse axonal injury can be observed in the pons using β-APP regardless of the causes, because many times, its presence in this encephalic structure only represents the delayed consequences of complex changes in cytoskeleton after trauma or ischemia, which leads to the collapse of anterograde axoplasmatic transport. This situation evolves to axonal swelling, varicosities and axonal bulbs and ends with axonal disruption, as it has been described before. Nevertheless, the topographic pattern of distribution of the expression of axonal injury along the structures of midline of the brain may be useful to diagnose the differences between traumatic and non-traumatic causes of axonal injury (²⁴)

On the other hand, the severity of the histopathological expression of diffuse axonal injury is assessed by the diameter of the swollen axons, which depends on several factors: the time elapsed since trauma, the diameter of the axons before swelling, length of time the secondary effect of trauma acts upon the axons and the time that any regeneration mechanism may take (²⁵).

Finally, the diffuse axonal injury can begin even 84 hours after the insult, which demonstrate the presence of a continuous secondary axonal damage. In addition, the immunoreactivity is still positive months after the insult, which confirms the persistence of the fast axonal flow. Anyway, this situation can give rise to degenerative diseases like Alzheimer or Parkinson. Fortunately, during the first moments after trauma there is a significant period of regeneration, this is why axonal injury can be healed, to a certain extent, in cases of mild trauma. (³⁹)

Another factor that makes some difference in the expression of β-APP is the diameter of axons. Its behavior in the optic chiasma and corpus callosum has been studied and the peak of expression in the former take place at 8 hours after trauma and at 28 hours in the latter, due to the thinner diameter of its axons and the higher proportion of microtubules of its cytoskeleton. The axons of the chiasma are thicker and have more density of neurofilaments. (⁴⁰).

Summary, the active process of β-APP formation and accumulation is clearly obvious in the first 24 hours after trauma, reaches a plateau at 10-14 days and then, began to disappear as gliosis grows around. Using immunohistochemistry techniques, the first changes in β-APP transient (swelling axons) were seen 30 minutes after the injury and they are visible during weeks. Thus, we can consider that it is positive in the first hour post-trauma, not only in the contused area, but in remote regions of the brain. The maximal peak of expression occurs 48 hours after injury and can be seen until two weeks after in white matter of both hemispheres, hippocampus, and cortex. In the brain stem it remains positive only for one week. (⁴¹,⁴²)

Once again, working under real work conditions restrict the interpretation of our results because it is no possible to know the intensity of the impact, and it is not practical to sample the whole brain. In the same way, this marker has some issues of sensibility (²⁴,⁴³,⁴⁴).

Nevertheless, our results are not very far from consensus, specially about the penumbra period in the first two hours after trauma. This period is an important medico-legal issue, above all in cases of mild head trauma. We were able to see weakly stained somas between 30 minutes and 2 hours post-injury. On the other hand, in cases of deferred death after head trauma (unique or repetitive), the pattern (''Z'' or diffuse) allows us to settle the relationship between the traumatic event and the cause of death, even though after 24 hours or more after the injury.