Spongiform encephalopathies

B lymphocytes and neuroinvasion

The latest episode in the long-running serial of scrapie pathogenesis appears on page 687 of this issue1, with a further molecular dissection of genetically altered mice. The Swiss team1 has taken advantage of the fact that the gene encoding the prion protein PrP — the leading candidate for the cause of all transmissible spongiform encephalopathies, including scrapie and Creutzfeldt-Jakob disease (CJD) — can be made inoperative (that is, ‘null’), without evident harm to the mice2. Working backwards from the target organ of disease (the brain) to peripheral (intraocular, intravenous or intraperitoneal) sites of infection, the authors have used surgical transplants in combination with normal and genetically altered mice in an effort to decode the pathogenesis of disease at the molecular level. They show that differentiated B lymphocytes are important for neuroinvasion — a finding with both public-health and therapeutic implications.

Intracerebral inoculation of the scrapie agent into a null mouse does not produce disease. However, intracerebral inoculation of a null mouse harbouring a small, PrP-expressing neural transplant destroys the graft, although the surrounding tissue is free from disease. In contrast, inoculation by peripheral routes does not destroy the graft: intraocular inoculation fails despite the fact that, in normal animals, the optic nerve is an efficient route of brain infection; and intraperitoneal or intravenous inoculation do not destroy the graft, even after lethal irradiation and transplantation of PrP-expressing haematopoietic stem cells (although these mice do develop infection of the spleen).

This series of experiments builds on a foundation of studies dating back to the 1960s, when Pattison and Millson established the widespread extraneural distribution of infectivity in scrapie-affected sheep and goats. The chief architects during the 1970s were Hadlow and Ecklund, followed, in the 1980s, by Fraser and Dickinson, and by Kimberlin and Walker. The importance of their contributions cannot be overemphasized in an era in which the power of molecular biology and accelerated publication has created a misguided tendency to neglect as already dated any research completed more than a few years earlier.

Thanks to them, and subsequent researchers (Kuroda, Diringer, Kitamoto and Lasmézas), we already know a good deal about the pathogenesis of transmissible spongiform encephalopathies. After introduction of the scrapie agent into various peripheral sites, including the stomach, infectivity first appears in the lymphoreticular system — tonsils, thymus, lymph nodes, and especially in spleen, in which a primary phase of replication occurs in B-cell-rich tissue fractions. Infectivity in these and other widely separated body organs implies that the infectious agent travels through the bloodstream, and this has been explicitly shown by isolation of infectivity from blood during both the preclinical and clinical stages of disease. Only after establishment of infectivity in peripheral organs does infectivity (and PrP) become detectable in the central nervous system; first in the spinal cord, and then in the lumbar and cervical cord segments and brain stem. Finally, genetically immunodeficient mice can partially resist intraperitoneal infection, yet they recover almost full susceptibility if immunologically reconstituted by the intraperitoneal injection of normal spleen cells.

This essential schema of scrapie pathogenesis is equally valid whether the infectious agent is a peculiar little virus or (as the Swedish Academy has just authorized) the prion protein. The observation by Klein et al.1, that mature B lymphocytes participate in neuroinvasion, should lead to an increasingly precise appreciation of the molecular basis of scrapie pathogenesis. For example, do B lymphocytes act as blood-borne carriers of the infectious agent, or do they act while in the spleen? Serial-infectivity measurements of blood lymphocytes and spleen during the incubation period could be compared in normal and B-cell-deficient mice. Blood infectivity could also be evaluated in splenectomized B-cell- and T-cell-deficient mice. What would happen after infection of a null mouse with PrP-expressing intracerebral and peripheral nerve implants, or intracerebral and splenic grafts? And could combinations of immunodeficient and PrP null mice strains be created?

Neuroinvasion may turn out to have a hierarchy of options. B lymphocytes contacting terminal nerve endings in the spleen could be the most efficient route; other lymphoid-tissue sites and transfer cells might be less efficient; and direct contact with nerve endings (short-circuiting the lympho-reticular system) least efficient of all. Physical transfer of the infectious agent from a lymphoid cell to a neuron might also require a ligand molecule — the mysterious ‘protein X’, for example. Among the candidates are two proteins with an affinity for PrP, which have been proposed as cell-membrane PrP receptors in this month's issue of Nature Medicine3,4.

What about the public-health and therapeutic implications of a role for B lymphocytes in peripherally acquired disease? There is a convincing link between the ‘new variant’ of Creutzfeldt-Jakob disease (vCJD) and consumption of infectious tissues from cattle with bovine spongiform encephalopathy (BSE). Moreover, the infectious agents of scrapie and CJD can be isolated from blood or blood components. The involvement of B lymphocytes in neuroinvasiveness adds another reason to consider a step that will deplete white blood cells during the commercial processing of blood, to reduce the possibility of transmitting CJD through blood products.

The outlook for therapeutic benefits is more problematic. Because we have no preclinical test for infection, nor any epidemiological clues as to who is at risk of vCJD, pre-emptive elimination of B lymphocytes is out of the question. PrP has been found in the tonsils of patients with vCJD, but this has so far been investigated only in clinically ill patients. We have no idea when — or even whether — PrP might appear during the incubation phase of disease. Moreover, even if PrP is present during the incubation period, we know that neuroinvasion occurs well before neurological symptoms. We cannot imagine that entire populations would submit to tonsillar biopsies as a screening procedure for vCJD. Thus, for the moment, we had best not push the implications of the involvement of B lymphocytes in neuroinvasion too far from the realm of basic science. When we know precisely how B lymphocytes influence neuroinvasion, it may be possible to nullify their effect without resorting to wholesale elimination strategies.


  1. 1

    Klein, M. A. et al. Nature 390, 687–690 (1997).

  2. 2

    Büeler, H. R. et al. Nature 356, 577–582 (1992).

  3. 3

    Martins, V. R. et al. Nature Med. 3, 1376–1382 (1997).

  4. 4

    Rieger, R., Edenhofer, F., Lasmézas, C. I. & Weiss, S. Nature Med. 3, 1383–1388 (1997).

Download references

Author information



Rights and permissions

Reprints and Permissions

About this article

Cite this article

Brown, P. B lymphocytes and neuroinvasion. Nature 390, 662–663 (1997). https://doi.org/10.1038/37733

Download citation

Further reading


By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.