3.1. Brain architecture was preserved both in PFA and Glyoxal fixation
It is important to maintain the macroscopic architecture and microenvironment of the brain samples by following the appropriate fixation technique. In the current study, we compared PFA (4%) with glyoxal (3%) fixation in maintaining brain structure in situ. Previous studies by Bussolati et al., 2017 and Richter et al., 2018 (13, 19) have reported that glyoxal has rapid penetration properties. However, an overnight incubation is necessary for fixing the deep brain structures. Also, in line with a previous study conducted by Iwashita et al., 2020 (20), in the present study it was noticed that mouse brain fixed with glyoxal exhibited lower stiffness than PFA (Figure. 1).
To evaluate and compare the preservation of brain anatomy across both PFA and glyoxal fixation conditions we used H&E staining. Both PFA and glyoxal were able to stain efficiently by H & E staining. As evident in Figure. 1 in whole coronal brain sections, brain structures, and morphology was well-preserved with each fixative.
3.2 Glyoxal fixation is optimal for obtaining consistent vascular staining in the brain.
To visualize blood vessels, immunohistochemical techniques targeting endothelial markers such as cluster of differentiation (CD) 31 (21, 22), basement membrane markers including laminin (23), and lectins that bind to the luminal part of the capillary endothelium through interactions with sugar residues (24) have been widely used. Prior reports (25–27) and from our lab experience, immunohistochemical vascular staining methods with 4% PFA fixed sections do not provide consistent results. This may be due to the antigen masking from the high crosslinking during fixation. To overcome this, different antigen retrieval methods can be used: including, high-temperature treatment in citrate buffer and enzymatic treatment with pepsin. However, this adds time and cost to the procedure. Further, this could also result in a non-uniform or inconsistent staining pattern due to differences in antigen exposure to retrieval methods.
Here we compared the 3 widely used endothelial markers Isolectin B4, CD31, and laminin for endothelial or vascular staining in PFA or glyoxal fixed free-floating brain sections. As shown in Figure. 2, we found that PFA fixed brain shows only IB4 staining while CD31 and laminin didn’t show any binding. However, from our experiments and as reported before, IB4 staining is consistent in PFA fixed sections. From our search, we don’t find any data showing that PFA fixation interferes with galactose residues in the cell membrane. This inconsistent staining may be due to differences in sample preparation or long tissue storage. Interestingly in glyoxal fixed brain sections, all three antibodies show consistent staining of endothelial cells/vessels.
3.3 Glyoxal fixation is better for higher quality blood-brain barrier/tight junction protein immunostaining.
The BBB restricts the movement of molecules between the blood and brain thereby maintaining cerebral homeostasis and proper neuronal function. BBB is composed of brain endothelial cells, microglia, pericytes, and a basement membrane (comprised of e.g. type IV collagen, laminin, and fibronectin), surrounded by astrocytes end-feet ensheathing (28). Leaky BBB is a serious concern in various CNS disorders like stroke, brain tumors, and neurodegenerative disorders (29). A high-resolution reliable staining method is warranted for detailed knowledge of BBB proteins, their interactions, and modifications as well as their distribution that maintains BBB integrity during normal physiology or disease.
Endothelial cells contribute to the core of the BBB (30–33). The barrier properties of these brain endothelial cells notably depend on tight junctions (TJ) between adjacent cells: TJs are dynamic structures consisting of several transmembranes and membrane-associated cytoplasmic proteins (34). Moreover, several studies have reported leaky BBB is associated with the loss of integrity in TJs (35). Here, we tested the efficacy of glyoxal fixation in staining the BBB proteins including tight junctions (TJs), transporter, and adherence junction. We tested the effectiveness of PFA and glyoxal fixation in visualizing three significant TJ proteins, ZO-1, Occludin, and Claudin-5 (36). None of these three antibodies showed specific binding in the PFA fixed brain sections (results not shown). However, for the glyoxal-fixed brain sections, continuous filaments of TJ proteins, ZO-1, occludin, and Claudin-5 are more easily detected (Fig. 3 ). This analysis suggests that the immunostainings performed after glyoxal fixation more readily allow the identification of TJ proteins. Moreover, we didn’t observe any difference with or without the use of blocking buffer indicating there is very little non-specific binding during this staining. We also tested the immunostaining of adherence junction protein VE-cadherin (37)., in the glyoxal fixed brain VE-cadherin staining specifically detects and localizes with the vascular staining. In PFA fixed samples VE-cadherin staining fails to consistently detect the antigen across the vasculature. Further, we compared the immunostaining for a BBB transporter Aquaporin-4 (38) and found strong vascular-specific staining for both PFA and glyoxal-fixed brain samples (Figure. 4)
A possible explanation for the lack of visualization of the TJ and adherence junction proteins in the PFA-fixed samples may be caused by high crosslinking and shrinking that mask the antigens. Interestingly, AQP4 staining was visualized in both PFA and glyoxal consistently. This can be related to the location of this protein. AQP4 is a water channel protein and is expressed in brain perivascular astrocyte processes (39). Together, direct exposure of endothelial cells to PFA while perfusion probably leads to vascular stiffness/shrinkage and masking endothelial cells proteins while this is not the case during glyoxal fixation where the brain is perfused with PBS and dissected into glyoxal solution.
3.4 Efficacy of PFA or glyoxal fixation in pericyte staining.
Having shown that glyoxal is a more effective fixative than PFA for vascular staining, we then investigated glyoxal’s efficiency in staining pericytes. Pericytes were found embedded within the basement membrane, both on straight sections and at branch points of capillaries, with projections extending from the soma to wrap around the underlying vessel. In recent times, various vascular functions of pericytes have been recognized including regulation of cerebral blood flow, maintenance of the blood-brain barrier (BBB), and control of vascular development and angiogenesis (40).
We used two common pericyte markers, NG2 proteoglycan, and PDGFRB (41), for the immunohistochemical staining for pericytes. Laminin was used as a vessel marker. The PFA fixed brain didn’t show any staining for pericyte markers, NG2, and PDGFRB (results not shown), while in glyoxal fixed brain shows uniform staining for pericyte markers along the vessels. We observed that NG2 immunohistochemistry showed flat staining around the vessels. Consistent with its receptor nature, PDGFRB staining was observed as more specific, spotty, and sporadic (Figure. 4)