Iron accumulation in the choroid plexus, ependymal cells and CNS parenchyma in a rat strain with low-grade haemolysis of fragile macrocytic red blood cells

Abstract

Iron accumulation in the CNS is associated with many neurological diseases via amplification of inflammation and neurodegeneration. However, experimental studies on iron overload are challenging, since rodents hardly accumulate brain iron in contrast to humans. Here, we studied LEWzizi rats, which present with elevated CNS iron loads, aiming to characterise choroid plexus, ependymal, CSF and CNS parenchymal iron loads in conjunction with altered blood iron parameters and, thus, signifying non-classical entry sites for iron into the CNS. Non-haem iron in formalin-fixed paraffin-embedded tissue was detected via DAB-enhanced Turnbull Blue stainings. CSF iron levels were determined via atomic absorption spectroscopy. Ferroportin and aquaporin-1 expression was visualised using immunohistochemistry. The analysis of red blood cell indices and serum/plasma parameters was based on automated measurements; the fragility of red blood cells was manually determined by the osmotic challenge. Compared with wild-type animals, LEWzizi rats showed strongly increased iron accumulation in choroid plexus epithelial cells as well as in ependymal cells of the ventricle lining. Concurrently, red blood cell macrocytosis, low-grade haemolysis and significant haemoglobin liberation from red blood cells were apparent in the peripheral blood of LEWzizi rats. Interestingly, elevated iron accumulation was also evident in kidney proximal tubules, which share similarities with the blood-CSF barrier. Our data underscore the importance of iron gateways into the CNS other than the classical route across microvessels in the CNS parenchyma. Our findings of pronounced choroid plexus iron overload in conjunction with peripheral iron overload and increased RBC fragility in LEWzizi rats may be seminal for future studies of human diseases, in which similar constellations are found.

Keywords: cerebrospinal fluid; erythrocyte osmotic fragility; iron overload; kidney proximal tubule; multiple sclerosis; zitter rat.

FIGURE 1

Iron density in the choroid plexus is strongly increased in LEWzizi rats. (A) Iron patterns (TBB staining) in the choroid plexus within the lateral ventricle (LV) and dorsal 3^rd ventricle (D3V) of 4‐months‐old (4M) Lewis and LEWzizi rats. Scale bars, 100 µm. (B) Total iron density of choroid plexus. Each data point represents the average of densitometric measurements of iron densities from lateral ventricles, dorsal 3^rd ventricles, 4^th ventricles and lateral recesses of the 4^th ventricle of an individual rat. AU, arbitrary units. (C) Densitometric measurements of iron density in the choroid plexus within the lateral ventricles (orange), dorsal 3^rd ventricles (blue), 4^th ventricles (magenta) and lateral recesses of the 4^th ventricle (green) in 4M Lewis and LEWzizi rats. AU, arbitrary units. (D) Detailed iron patterns of a choroid plexus in the lateral ventricle of a 4M LEWzizi rat. Scale bar, 50 µm. (E) Ratio of albumin concentrations (Q_Alb) between paired CSF and blood plasma samples taken from 4M Lewis and LEWzizi rats. (F) Iron levels within the cerebrospinal fluid (CSF) of 4M Lewis and LEWzizi rats. (G and H) Densitometric quantification of ferroportin (G; FPN) and aquaporin‐1 (H; AQP1) expression in choroid plexuses within the lateral ventricles of 4M Lewis and LEWzizi rats. AU, arbitrary units. (I and J) Apical expression of ferroportin (I) and aquaporin‐1 (J) in choroid plexus epithelial cells. Representative pictures were taken from choroid plexuses in lateral ventricles of 4M Lewis and LEWzizi rats. Scale bars, 50 µm. (B, E‐H) Reported statistics results from unpaired, two‐tailed Student's t‐tests. (C) Reported statistics result from one‐way ANOVAs. (b, c, e‐g) Dots represent individual rats; error bar ± SD; *p < 0.05; **p < 0.01; ****p < 0.0001; ns, not significant

FIGURE 2

Numbers of iron‐positive ependymal cells are increased in LEWzizi rats. (A) Two major iron patterns were detected in the ependymal ventricular lining: vesicular iron positivity and diffuse cytoplasmic iron accumulation. Representative pictures of DAB‐enhanced TBB stainings were taken from the lateral ventricle of a 4M LEWzizi rat. Scale bars, 10 µm. (B) Percentage of iron‐positive ependymal cells lining different borders of the lateral ventricle of 4M Lewis and LEWzizi rats as shown in the schematic hemispheric cross‐section. Iron staining patterns were categorized as either vesicular or diffuse and were determined separately for ependymal cells proximal to the lateral septal nucleus (blue), corpus callosum (magenta) or caudoputamen/striatum (green). Cx, cortex; CC, corpus callosum; CP, caudoputamen; LV, lateral ventricle. (C‐F) Percentage of iron‐positive ependymal cells lining the lateral ventricle (C), dorsal 3^rd ventricle (D), aqueduct (E) or central canal of the spinal cord (F). Data represent the percentage per ventricular compartment independent of the surrounding parenchyma of 2M, 4M and 8M Lewis and LEWzizi rats. Per individual rat and tissue section, between 225 and 1669 (lateral ventricle), 44 and 274 (dorsal 3^rd ventricle), 58 and 634 (aqueduct) and 26 and 121 (central canal of the spinal cord) total ependymal cells were counted, of which numbers of iron‐positive cells were registered as well. (G) Predominant basolateral expression of ferroportin in ependymal cells lining the lateral ventricles of 4M Lewis and LEWzizi rats. Scale bars, 25 µm. (H) Densitometric quantification of ferroportin (FPN) expression in the ependymal linings of the lateral ventricles (LV), dorsal 3^rd ventricle (D3V) and aqueduct (AD) of 4M Lewis and LEWzizi rats. AU, arbitrary units. (B) Reported statistics result from one‐way ANOVAs followed by Tukey's multiple comparisons tests. (C‐F; H) Reported statistics result from unpaired, two‐tailed Student's t‐tests. (B, C‐F) Dots represent individual rats; error bar ± SD; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant

FIGURE 3

Abnormally high iron accumulation in the brain of LEWzizi rats. (A and B) DAB‐enhanced TBB staining (brown) for the detection of non‐haem iron in representative hemispheric brain (A) and spinal cord (B) cross‐sections of 4‐months‐old (4M) Lewis and LEWzizi rats. Sections were counterstained for nuclei (blue). Numbers refer to regions of interest quantified in (C). (A) Scale bar, 1 mm; (B) scale bar, 500 µm. (C) Densitometric measurements of iron density in the striatum, mesencephalon, medulla oblongata and cervical spinal cord of 2M, 4M and 8M Lewis and LEWzizi rats. Numbers refer to areas of interest in (A), where pictures were taken under standardised conditions. Reported statistics result from unpaired, two‐tailed Student's t‐tests. Dots represent individual rats; error bar ± SD; AU, arbitrary units; *p < 0.05; **p < 0.01; ***p < 0.001; ns, not significant. (D) Iron patterns visualised by TBB staining in the mesencephalon of 4M Lewis and LEWzizi rats. Scale bar, 20 µm

FIGURE 4

Low‐grade intravascular haemolysis of fragile macrocytic red blood cells in LEWzizi rats. (A‐D, G‐I) Complete blood counts of K₃EDTA‐treated blood of 4M Lewis and LEWzizi rats. RBC, red blood cell; MCV, mean cell volume of RBCs; RDW, RBC distribution width; MCH, mean cellular haemoglobin content of RBCs; HDW, haemoglobin distribution width. (E and F) Folate (E) and vitamin B12 (F) levels in the serum of 4M Lewis and LEWzizi rats. (J) Reticulocyte counts in K₃EDTA‐treated blood of 4M Lewis and LEWzizi rats. (K) Osmotic fragility of RBCs upon hypotonic challenge. Depicted is the NaCl concentration leading to 50% haemolysis of RBCs within a blood sample. (L‐N, P, Q) Automated measurements of free haemoglobin (L), cell‐free iron (M), lactate dehydrogenase (LDH, N), haptoglobin (P) and potassium (Q) levels in the plasma of 4M Lewis and LEWzizi rats. (O) Soluble CD163 (sCD163) levels in the serum of 4M Lewis and LEWzizi rats determined by ELISA. Reported statistics result from unpaired, two‐tailed Student's t‐tests. Dots represent individual rats; error bar ± SD; *p < 0.05; **p < 0.01; ***p < 0.001; ****p < 0.0001; ns, not significant

FIGURE 5

Kidney proximal tubules show increased iron accumulation in LEWzizi rats. (A‐E) Iron quantification via the ferrozene method of liver (A), lung (B), heart (C), spleen (D) and kidney (E) homogenates derived from 4M Lewis and LEWzizi rats. Animals were thoroughly perfused with PBS prior to tissue dissection. Reported statistics result from unpaired, two‐tailed Student's t‐tests. Dots represent individual rats; error bar ± SD; *p < 0.05; **p < 0.01; ns, not significant. (F) Iron staining patterns in kidneys of 4M Lewis and LEWzizi rats. Boxed areas in the cross‐sections are shown in higher magnification as well as depicting iron accumulation in renal proximal tubules. Scale bars cross‐sections, 1 mm; scale bars magnification, 100 µm