Architecture of secondary lymphoid tissue in sheep experimentally challenged with scrapie

Abstract

Scrapie is a transmissible spongiform encephalopathy in which there is an accumulation of the abnormal form of the prion protein, PrP^sc, in the lymphoreticular system and nervous system. There is a particular accumulation of PrP^sc on follicular dendritic cells within the germinal centre of B-cell follicles. Because accumulation of PrP^sc in the nervous system leads to neuronal cell loss we have examined PrP^sc accumulation in the prescapular and mesenteric lymph nodes in relation to lymph node architecture of scrapie-challenged sheep. We demonstrate that an accumulation of PrP^sc in the lymph node fails to result in gross defects in the microanatomy and phenotype of T- and B-cell areas in the lymph nodes.

Introduction

Scrapie is a naturally occurring transmissible spongiform encephalopathy (TSE) of sheep and goats and is a prototypic disease for Creutzfeldt–Jakob disease in humans and bovine spongiform encephalopathy (BSE) in cattle. The pathology of such diseases is characterized by vacuolation, neuronal loss and glial cell activation/proliferation. The exact nature of the transmissible agent remains controversial, although many experiments support the ‘protein-only’ hypothesis, which postulates that the agent is devoid of nucleic acid and consists of an abnormal conformer (PrP^sc) of the host encoded prion protein (PrP^c).¹ In this hypothesis, PrP^sc replicates by converting PrP^c into more PrP^sc with this conformational change postulated to be a post-translational event. However, not all researchers agree with this aetiology.² Prion replication requires PrP^c because mice deficient of PrP^c are unable to accumulate prion infectivity,³ while disease is exacerbated in mice over-expressing PrP^c.⁴

Ovine susceptibility to scrapie is partially controlled by the PrP gene.⁵ The main polymorphic PrP gene sites associated with susceptibility or resistance to scrapie are codons 136 (A or V), 154 (R or H) and 171 (R, Q or H). Although there are breed differences, in general the ARR genotype is associated with resistance while the VRQ and ARQ genotypes are associated with susceptibility to disease.

In natural and experimental scrapie, the abnormal form of the prion protein (PrP^sc) is often found to sequentially accumulate first in the lymphoreticular system, then the peripheral nervous system before invading the central nervous system.⁶ This pathogenesis is PrP genotype dependent. Heterozygote genotypes have less involvement of peripheral tissues in PrP^sc deposition.^7–9 In VRQ/VRQ sheep, PrP^sc accumulation has been observed in the gut associated lymphoid tissue as early as 2–3 months of life in natural scrapie and 3–5 weeks after experimental oral challenge in scrapie susceptible sheep.^7,10

There is evidence that both the peripheral nervous system and the lymphoreticular system (LRS) can be involved in the development of scrapie.¹¹ High-dose peripheral infection might cause direct neuroinvasion via the vagus nerve, whereas a low-dose infection may be amplified in the LRS prior to neuroinvasion.^12–14 The role of the LRS in the development of disease is unclear although murine models of scrapie indicate that both the follicular dendritic cells (FDC) and B cells may play a role in the pathogenesis of scrapie. FDCs show accumulation of the abnormal PrP^sc during disease and may be important for prion replication,¹⁵ while B cells need not express PrP^c but function by inducing the development of PrP-positive FDCs.¹⁶ In a recent paper¹⁷ treatment with LTβR-Ig to remove FDC 2 weeks after oral scrapie inoculation had no effect on incubation time. However, treatment with LTβR-Ig prior to oral scrapie inoculation blocked accumulation of PrP^sc in the Peyer's patch and mesenteric lymph nodes and prevented neuroinvasion. Thus, it appears that infectious prions can pass from the alimentary canal, the probable portal of entry, and associate with FDCs in gastrointestinal associated lymph nodes where infection is enhanced prior to invasion of the central nervous system (CNS) and development of clinical disease.

In the CNS, accumulation of PrP^sc is associated with the characteristic destruction of neuronal cell bodies and development of a progressive spongiform encephalopathy. Similarly, if accumulation of PrP^sc on the FDC network were to impair the function of FDCs then this could have consequences for the development of B-cell memory, which requires an intact FDC network, and proper development of the germinal centre. It has previously been reported that PrP^sc accumulation in lymph nodes of mice infected with the ME7 strain of scrapie show hyperplasia of FDCs.¹⁸ However, no-one has previously reported microscopic alterations in lymph node architecture concerning cell populations other than FDCs in scrapie-infected sheep. In this paper we have used monoclonal antibodies to sheep lymphocyte subsets in flow cytometry and immunohistochemistry to determine if there are changes in the cellular phenotype and normal cellular architecture of the germinal centres in PrP^sc positive LRS tissues. We report that accumulation of PrP^sc in sheep secondary lymphoid tissue, following experimental challenge of sheep with scrapie brain homogenate, fails to result in gross defects in the microanatomy of T- and B-cell areas in lymph nodes.

Materials and methods

Experimental animals

Twenty New Zealand Poll Dorset sheep, supplied from the DEFRA SF (scrapie free) flock, were inoculated subcutaneously in the lateral side of the neck with 2 ml of a 10% sheep scrapie brain pool homogenate, SSBP/1, as described previously.⁹ Seven New Zealand Poll Dorset sheep were kept as controls. Sheep ranged in age from 6 months to 2 years at the time of inoculation. Details of animals used are presented in Table 1. Control and scrapie inoculated sheep were housed separately.

Table 1.

Details of animals recruited to the study

				‡PrPsc detection
Status	Genotype	Sex	Incubation period/ survival time	PSLN	MLN
*Clinical + PrPsc
″H140	VRQ/VRQ	F	133	+	±
″H131	VRQ/VRQ	F	162	+	−
″H175	VRQ/VRQ	M	163	+	+
″H78	VRQ/VRQ	M	169	+	+
″H37	VRQ/ARQ	M	202	+ +	±
″H31	VRQ/ARQ	M	195	+	±
″H10	VRQ/ARQ	F	204	+ +	±
″980749	VRQ/ARQ	M	226	+ +	−
†Clinical
″H7	VRQ/ARQ	F	182	−	±
″970599	VRQ/ARQ	M	212	−	−
″980624	VRQ/ARQ	M	217	−	−
″F159	VRQ/ARR	M	197	−	−
″F321	VRQ/ARR	M	190	−	−
″F312	VRQ/ARR	M	201	−	−
″970613	VRQ/ARR	M	246	−	−
″971158	VRQ/ARR	M	242	−	−
″980164	VRQ/ARR	M	241	−	−
Non clincal
″D258	ARR/ARR	M	270	−	−
″980213	ARR/ARR	M	270	−	−
″970648	ARR/ARR	M	270	−	−
Controls
″970616C	VRQ/ARQ	M		−	−
″980741C	VRQ/ARQ	M		NT	NT
″980757C	VRQ/ARQ	M		NT	NT
″980218C	VRQ/ARR	M		−	−
″971174C	VRQ/ARR	M		−	−
″971213C	VRQ/ARR	M		NT	NT
″970623C	ARR/ARR	M		−	−

^*Following challenge with scrapie brain homogenate sheep developed clinical signs of scrapie with accumulation of PrP^sc in the lymph nodes.

^†Following challenge with scrapie brain homogenate sheep developed clinical signs of scrapie without accumulation of PrP^sc in the lymph nodes.

^‡The percentage of lymphoid follicles that showed PrP^sc staining was determined for prescapular lymph node (PSLN) and mesenteric lymph node (MLN) sections. −, no PrP^sc positive follicles; ±, 0–25% PrP^sc positive follicles; +, 26–50% PrP^sc positive follicles; + +, 51–75% PrP^sc positive follicles.

Tissue collection

Experimentally infected animals (clinical and non-clinical) and negative controls were killed by intravenous injection of pentobarbitol. A diagnosis of scrapie was made by histopathological examination of the brain or Western blotting of proteinase K-digested PrP^sc extracted from the brain. Tissue samples were taken from the prescapular lymph node (PSLN) draining the site of inoculation and mesenteric lymph node (MLN) and either (i) fixed in neutral buffered formalin (ii) immediately frozen in OCT embedding solution using isopentane and liquid nitrogen, or (iii) collected into sterile RPMI medium/10% fetal bovine serum. Frozen tissue blocks were wrapped in aluminium foil and stored at −70° until tested.

PrP^sc detection

Formalin fixed tissue blocks were treated with 98% formic acid for 1 hr before processing, washed, dehydrated by routine methods and embedded in paraffin wax. Sections 5 µm thick were cut, air dried overnight, dewaxed and rehydrated. Sections were immersed in distilled water and heated in a microwave oven at 100° for 20 min and washed with phosphate buffered saline. Endogenous peroxidase was blocked with 1% hydrogen peroxide in methanol and epitopes revealed by incubation with 0·1% trypsin for 20 min at room temperature. Non-specific binding was blocked with 1% normal horse serum. Sections were incubated overnight with anti-PrP antibody, FH11, for 1 hr with biotinylated horse anti-mouse immunoglobulin G (IgG). Signal was amplified by incubating the sections for 30 min in VECTASTAIN^® ABC Reagent; an Avidin/Biotin Complex Reagent conjugated to peroxidase (VECTASTAIN^® ABC KIT, Vector Laboratories, Burlinghame, CA). Sections were developed with VECTOR^® NovaRED (Vector Laboratories) and counter-stained with haematoxylin. Between steps sections were washed three times in phosphate-buffered saline (PBS).

Lymphocyte phenotyping

Cryostat sections of frozen tissue were cut at 5 µm and air dried. Sections were fixed briefly with acetone and non-specific binding was blocked with horse serum (Vectastain kit). Sections were incubated with monoclonal antibodies to pan B (DU2104), CD21 (DU254), IgM (VPM13), CD4 (SBUT4), CD2 (36F), CD8 (SBUT8), CD45 (TD14), major histocompatibility complex (MHC) class II (VPM54) and CD14 (VPM65) overnight at 4°. All primary antibodies were undiluted culture supernatants. Primary antibody binding was detected using peroxidase conjugated goat anti-mouse IgG or IgM secondary antibodies (Vectastatin kit). Signal was amplified with avidin biotin complex reagent conjugated to horseradish peroxidase (ABC–HRP). Sections were developed with 3,3′-diaminobenzidine (DAB) and counterstained with Mayer's haematoxylin before dehydrating and mounting. Endogenous peroxidase was blocked with 0·03% hydrogen peroxidase in PBS. Between each step sections were thoroughly washed with PBS. Omission of primary antibody was used to provide a negative control.

Flow cytometry

Lymphoid tissue was teased apart with forceps, passed through a sieve and lymphocytes isolated using Ficoll density gradient centrifugation. Cells were stored in liquid nitrogen until required. Lymphocytes were incubated with 20% normal rabbit serum in PBS to block non-specific binding then stained with primary antibodies to pan B (DU2104), CD21 (DU254), IgM (VPM13), CD1b (CC20), CD2 (36F), CD4 (SBUT4), CD8 (SBUT8), MHC class II (DR) (VPM54) and CD14 (VPM65). Cells were incubated with FITC conjugated goat anti-mouse immunoglobulin of relevant isotype (Caltag, Northampton, UK) for 30 min on ice. Unconjugated isotype-matched irrelevant control Mabs were included in all experiments. Samples were analysed on a FACSORT flow cytometer using CellQuest software (Becton Dickinson, San Jose, CA). For each sample 10 000 events were collected using the light scatter profile to gate the lymphocyte cell population. One way analyses of variance using Kruksal–Wallis test for nonparametric data were used to compare the expression of lymphocyte markers between the four groups of animals for each tissue type and antibody tested. A P-value < 0·05% was considered significant.

Results

Scrapie-challenged sheep

All of the Poll Dorset sheep carrying the VRQ allele and challenged with scrapie brain homogenate (SSBP/1) developed clinical signs of scrapie (Table 1). Animals included VRQ/VRQ homozygotes as well as VRQ heterozygotes carrying either the ARQ or ARR allele. Incubation times were shorter in animals homozygous for the VRQ allele compared with heterozygotes. Animals homozygous for the ARR genotype failed to develop clinical signs of scrapie following challenge.

PrP^sc detection

In scrapie-susceptible sheep with the VRQ PrP allele accumulation of PrP^sc in the lymph nodes was not detected in all animals even though all animals showed clinical signs of scrapie (Table 1). Based on the appearance of clinical signs of scrapie and an accumulation of PrP^sc in the lymph nodes scrapie-susceptible sheep could be divided into two groups: those with clinical signs of scrapie and PrP^sc deposition in the lymph nodes (VRQ/VRQ sheep and 4/7 VRQ/ARQ sheep) and those with clinical signs of scrapie but no detectable PrP^sc deposition in the lymph nodes (3/7 VRQ/ARQ and VRQ/ARR sheep). This is in agreement with previously reported data.⁹

Where PrP^sc was detected labelling appeared as a reticular network pattern within the basal light zone of B-cell lymphoid follicles (Fig. 1). The prescapular lymph nodes taken from sheep with the VRQ/ARQ genotype and clinical signs of disease showed the highest number of follicles staining for PrP^sc. However, in both VRQ/VRQ and VRQ/ARQ sheep, on average only two to three follicles per section showed PrP^sc accumulation with the staining being relatively weak. In one VRQ homozygote and one VRQ/ARQ animal PrP^sc deposition was found in the PSLN but not in the MLN. PrP^sc was not detected in the lymph nodes of scrapie-challenged animals with the ARR/ARR scrapie-resistant PrP genotype or control animals.

Figure 1.

PrP^sc accumulation in the PSLN of a VRQ/ARQ scrapie-challenged sheep.

Lymphocyte phenotyping

Frozen tissue sections from scrapie-susceptible (those encoding the VRQ allele) and -resistant sheep stained with Mayer's haematoxylin solution showed no apparent morphological alterations to the B-cell follicles. Optimum immunolabelling was obtained by incubating sections with the primary antibody overnight at 4° and blocking endogenous peroxidase between the secondary antibody step and the ABC–HRP step. The pattern of staining was similar in PSLN and MLN tissue for all antibodies tested and there was no obvious difference between sheep with and without clinical signs of scrapie or sheep with and without PrP^sc deposition in the lymph nodes.

Strong labelling of CD21 epitopes, using DU254 monoclonal antibody (mAb), was present in the light zone and mantle zone of the lymphoid follicle (Fig. 2). IgM-positive cells, identified with VPM13 mAb, were demonstrated in the CD21 positive areas. The follicles contained mainly B cells, with the T cells principally demonstrated in the paracortex of the lymph node immediately adjacent to follicles. However, a few CD4 and CD8 T cells were found scattered throughout the follicle. CD14 epitopes, detected with VPM65, were scattered through the lymph node although immunolabelling was weak. MHC class II antigens, identified by VPM54, were found within the follicle as well as in the paracortex. Labelling with TD14, corresponding to CD45, identified intense staining within the follicle but also in the paracortex.

Figure 2.

Immune phenotyping on lymph nodes using (a) pan B, (b) CD21, (c) IgM, (d) MHC class II DR, (e) CD2 (f) CD4, and (g) CD8 monoclonal antibodies. The pattern of staining was similar in PSLN and MLN tissue and there was no obvious difference between sheep with and without clinical signs of scrapie or sheep with and without PrP^sc deposition in the lymph nodes.

To confirm the immunohistochemistry results lymphocytes isolated from tissues were also analysed by flow cytometry. There was no significant difference in the expression of lymphocyte surface CD markers between the four groups of animals for each antibody tested (Fig. 3).

Figure 3.

The percentage of cell subsets found in (a) PSLN (b) MLN and (c) PBL. Cells were isolated from tissues or blood and analysed by flow cytometry. Using Kruksal–Wallis test there was no significant difference in the expression of lymphocyte surface CD markers between the four groups of animals for each antibody tested. Animals showing clinical signs of scrapie and accumulation of PrP^sc in the lymph nodes, animals showing clinical signs of scrapie without accumulation of PrP^sc in the lymph nodes, □ animals inoculated with PrP^sc but failed to develop clinical signs of scrapie or PrP^sc accumulation in lymph nodes, age matched control sheep.

Discussion

Sheep that developed clinical signs of scrapie also showed brain histology associated with scrapie. Classical clinical signs of scrapie were observed including pruritus and loss of coordination. In general, as previously reported, incubation periods were slightly shorter in Poll Dorsets compared to Cheviots.

In this present study VRQ/VRQ sheep showed similar PrP^sc deposition in both prescapular and mesenteric lymph nodes. This is in contrast to Cheviots experimentally challenged in the same manner as the Poll Dorset sheep used in this study where PrP^sc deposition was more intense in prescapular than mesenteric lymph nodes.⁹ In comparison with the Cheviots, immunolabelling of lymphoid tissues in Poll Dorset sheep was generally weaker. This agrees well with published data showing weaker immunolabelling of PrP^sc on brain sections in Poll Dorset sheep.¹⁹

In contrast to Cheviots, PrP^sc deposition was not found in PSLN of all VRQ/ARQ Poll Dorset sheep and immunolabelling was weaker in the MLN of Poll Dorset sheep. The difference in number of follicles staining for PrP^sc between PSLN and MLN in the present study may be due to accumulation of PrP^sc in a lymph node closest to site of inoculum. In natural scrapie the number of follicles staining for PrP^sc was similar in PSLN and MLN.^7,8 The difference in accumulation of PrP^sc in lymphoid tissue between natural and experimentally challenged scrapie may be due to route of exposure, the scrapie strain used for infection or the length of the clinical period. Experimentally infected scrapie cases were killed early in the clinical course in accordance with humane end points agreed with the Home Office.

PrP^sc deposition was not observed in lymph nodes taken from VRQ/ARR Poll Dorset sheep with clinical signs of scrapie and accumulation of PrP^sc in the brain while weak staining was recorded in some Cheviots.⁹ However, in the rare cases of natural scrapie in VRQ/ARR sheep PrP^sc was not found in the LRS.^8,20,21 This suggests that VRQ/ARR scrapie sheep are less capable of supporting peripheral accumulation of PrP^sc in the LRS. Accumulation of scrapie agent or PrP^sc in the LRS is not essential for disease development in scrapie-affected sheep^7,8 and hence it is possible that direct neuroinvasion without prior replication in the LRS can occur in sheep with susceptible genotypes.

As with already published data PrP^sc deposition appeared as a reticular pattern in the light zone of B-cell follicles.⁸ This reticular network staining pattern is suggestive of labelling of follicular dendritic cells. In addition, an association between PrP^sc accumulation and the FDC network in lymphoid follicles in the spleen and lymph node in scrapie- and BSE-challenged mice has clearly been demonstrated.²² Similarly, it has been reported that PrP^sc does not stain the cytoplasm of B or T cells whereas PrP^sc can be found to accumulate within the cytoplasm of a FDC.²¹

We found that the pattern of immunostaining for lymphocytes was the same in PSLN and MLN and that there was no difference between sheep with and without clinical scrapie or accumulation of PrP^sc in the lymph nodes. The pattern of staining was similar to that observed in lymph tissues from healthy sheep.²³ The results obtained from flow cytometry analysis of lymphocytes isolated from the nodes is in accordance with the immunohistochemistry results. Therefore, the current work shows that deposition of PrP^sc on FDC does not result in disruption of germinal centre architecture and/or function. However, the relatively low levels of PrP^sc accumulation in the lymphoid tissues of these experimentally infected sheep may be insufficient to cause major disruption of germinal centre function.

In human immunodeficiency virus (HIV) the virus is reported to be trapped by FDCs²⁴ with the destruction of the B-cell germinal centre network and collapse of the normal FDC architecture.²⁵ Hence accumulation of HIV on FDCs contributes to HIV pathogenesis although in HIV B-cell function is also modulated separately.²⁶ The slow accumulation of PrP^sc on FDCs and the slow turnover of these cells could therefore similarly result in the destruction of FDCs and interfere with B-cell differentiation in prion diseases. Hence B cell function may be a good indicator of disease progression indicating destruction of FDCs. However, neither B cells nor immune complexes are required for neuroinvasion from the periphery.²⁷ There is evidence that complement plays a role in the association between PrP^sc and FDC in mice^28,29 but since antibody–antigen complexes are not required for PrP^sc aggregation, PrP^sc may somehow exploit antibody-independent mechanisms for PrP^sc accumulation on FDCs that do not trigger an immune response or cause pathogenic effects.

In conclusion, there is no evidence from flow cytometry and immunohistochemistry that PrP^sc accumulation on FDCs results in any major change to immune cell profiles or germinal centre architecture. This suggests that this PrP^sc accumulation may be analogous to immune complex localization and does not result in cell death as happens with cells in the CNS.