Immunology Today M a y 1984

Ataxia-telangiectasia

I m m u n e dysfunction is one of many defects from Richard Gatti Ataxia-telangiectasia (AT), a progressive and uniformly fatal inherited neuroimmunological disorder, affects approximately I in 40 000 children. Patients are born to normal parents and appear to develop normally for about two years. They t h e n b e g i n to stagger (ataxia) a n d show signs of degenerating cerebellar function; by five years of age, they have dilated blood vessels (telangiectases) over the exposed bulbar conjunctiva and skin of the ear. By 10 years of age, they are usually confined to a wheel-chair. Most AT patients have IgA and IgG2 deficiency as well as various T cell-associated i m m u n e dysfunctions. Because one of every five patients develops cancer, usually lymphoid, d u r i n g their shortened life-span, it is believed that u n r a v e l l i n g the pathogenesis of A T will also shed light upon the genetics of cancer susceptibility and upon the relationship of immunodeficiency to oncogenesis. A recent conference* focused on four areas of this complex disorder: DNA repair/replication, genetics, neuropathology and immunopathology. An early clinical observation that A T patients are unusually sensitive to X-rays led to the demonstration that fibroblast strains from these patients are about three times more sensitive to killing by ionizing radiation than are the corresponding cells from normal individuals. This unique sensitivity to X-rays and )'-rays, but not to short wavelength uv light, is now a hallmark of the disease. Thus far, there has been no unequivocal demonstration of a defect in any particular step in a DNA repair pathway. DNA strand breaks produced by X-rays are rejoined normally in A T cells and the cells are not sensitive to some other agents, such as methylmethane sulfonate, that produce strand breaks. R. Painter (Univ. of California, San Francisco) has found that chain elongation of DNA during synthesis in A T cells is unaffected by as much as 5 000 rads of X-rays. The deficiency could be in a regulatory process which normally delays synthesis on the damaged D N A templates until they have been repaired.

*A Kroc Foundation conference held on 16-20 January 1984 at Solvang, California. Ray A. Kroc, the founder of the foundation and of McDonald's food chain, died on 14January 1984.

Y. Shiloh (Children's Hospital, Boston) described a sensitivity of A T cells to bleomycin and other agents, such as neocarzinostatin and hydrogen peroxide, which have in common with

X-rays the property of generating free radicals that attack the deoxyribose sugar moiety. Shiloh suggested that AT cells may be deficient in repair of only specific types of D N A lesions, such as those that result in a gap with a 3' phosphate end which cannot be ligated unless the phosphate group is first removed. C. Arlett (Univ. of Sussex, Brighton, UK) reported that untransformed AT fibroblasts are hypomutable for the production of 6-thioguanine resistance to ),-rays so that the putative translesion DNA synthesis is evidently not error prone; it is likely that the lesion simply results in lethality. Curiously, Arlett also finds that when an A T fibroblast strain is transformed with SV40 virus, it becomes mutable and less sensitive to y-rays. M. Paterson (Chalk River Nuclear Labs, Ontario, Canada) suggested, by analogy with properties of recA mutants in bacteria and rad52 mutants in yeast, that the primary defect in A T could be an enzyme involved in genetic recombination. Continued on p. 122

The biochemistry and genetics of complement

Agreement n o w the norm from Michael Hobart Many immunologists hold that complement is baffling or irrelevant or, most conveniently, both but a recent meeting* emphasized that complement is interesting and that it may be important, even if only as an elegant model system. Moreover agreement is now the norm. The biochemistry has now advanced to the point where the outlines of the reaction mechanisms are agreed and work is under way with detailed analysis and model building. The latter approach (M. Colomb, Grenoble; J. Fothergill, Aberdeen; and J. Gagnon, Oxford) provides stimulating questions for the sequencers (themselves included), though unfortunately not yet for crystallographers. It is now agreed that the early components of the complement pathways are serine proteases with peculiar properties of structure and * Held at the Royal Society, London, on 25 and 26 January 1984.

specificity. They fall into fanfilies: C l r with Cls, perhaps distantly related to factor D, and C2 with factor B. Their substrates are other complement components and the control of their actions is subtle. We still need a fully satisfying account of how C 1r is able to effectively cleave C 1s into its fully active form only when C l q interacts with a suitable molecule (normally fixed antibodies). C 1s can cleave the dissimilar molecules C4 and C2 but not their close relatives C3 and ~~i~.~

Continued on p. 122

© 1984,ElsevierSciencePublishersB.V., Amsterd~ 0167 4919/84/$02.00

122 Ataxia-telangiectasia: continuedfrom p. 121 in the laboratories of Painter and Attempted replication Over a lesion that Jaspers. As W. Salser (UCLA) pointed constituted a strand break should result out, the lingering hypothesis whereby a topologically in a chromosomal break large deletion of adjacent genes might and this could account for the chromo- ~expl~'p the pleomorphism of AT pathsomal fragility that is characteristic of 01ogy co~a]d be dismissed on the basis of AT cells. Although the diminished inthis evidence. Although it has not yet hibition of DNA synthesis in irradiated been possible to correct the in-vitro A T AT cells suggests that the primary defect defect of radiation resistance by transinvolves an abnormality in DNA replicafecting the fibroblasts with DNA from tion, N. G. J. Jaspers (Erasmus Uniindividual normal, chromosomes, this versity, Rotterdam) pointed out that AT seemed to some a promising approach to cells also respond untowardly to X-rays the localization of the AT gene(s) in the in the absence of DNA synthesis in that human genome. D. Botstein (MIT, increased frequencies of chromosomal Cambridge) pointed out the potential aberrations are observed after radiation difficulties of genetic mapping by exposure to AT cells in the G2,phase. restriction fragment length polymorphIt was generally agreed that published isms (RFLPs) in this disorder, considerdata of in-vitro post-radiation survival ing its autosomal recessive nature, the curves showed significant: differences lack, thus far, of a reliable method for between controls and obligatory AT heterozygote identification and the probheterozygotes as groups; however, no ability that the AT phenotype may single assay appeared to discriminate reflect as many as a half-dozen different individual heterozygotes from nongenetic addresses. carriers sufficiently well to be used in Within the confines of the above genetic counselling. Paterson presented reservations, linkage data were predata showing good separation of 10 sented which suggest that the AT gene is heterozygotes from controls when he not closely linked to the Gm loci in the exposed fibroblasts from these patients to IgH complex on chromosome 14q32, an chronic, low-dose irradiation. Perhaps area where break points for tandem this assay, or combination of assays, will chromosomal translocations have been prove sufficiently reliable to distinguish d e s c r i b e d in A T patients. AlphaAT heterozygotes within families, if not fetoprotein (AFP) is raised in all but a few within populations. The time has come AT patients; those who have normal for 'blind' samples to be submitted to levels are suspected of belonging to a those laboratories confident that their separate subset of the disorder. Informaassay separates A T heterozygotes from tion regarding the cloning and mapping controls. of the AFP gene was presented by A. Dugaiczyk (Univ. of California, RiverGenetics side). M. Swift (Univ. of N. Carolina, The neuropathology session was Chapel Hill) presented a preliminary unique in being the first of its kind which analysis of an on-going epidemiological specifically addressed AT before a multistudy of American families with AT. disciplinary audience with expertise in Despite a very low incidence of conmolecular biology, genetics, immunsanguinity in these American families, ology and medicine. H. Vinters (Univ. the autosomal recessive pattern of of Western Ontario, London) presented inheritance was supported by an work performed at U C L A which showed incidence of affected members approxithat because basket cells represent a footmating 30%. Cancer incidence among print of Purkinje cells and were present 211 patients was 19%, although more in normal numbers in the cerebellum of than half these patients are still alive and AT patients, normal numbers of Purcancer incidence among those dead was kinje cells must have existed previously 36%. The data also suggest an increased in these patients and deteriorated cancer incidence among A T heteroprogressively with time. This picture zygotes and a carrier frequency as high as correlates well with the progressive 1:100; this would make AT the most nature of the ataxia. P. Rakic (Yale common cancer susceptibility gene in the Univ.) noted that Vinters' ectopic general population. Purkinje cells in the molecular layer of Before strategies for mapping and the cerebellum would have had to arrive eventual cloning of the AT gene could there by the thirteenth week of gestation, considered, the issue of complementasuggesting that the genetic disorder tion groups was once again raised. expresses itself much earlier than had Recent fusion experiments between 10 been previously appreciated. Since A T fibroblast strains identify at least four Purkinje cells do not divide after that complementation groups. These results time, the cause of their disappearance in derived from independent experiments Continuedon p. 123

Immunology Today, vol. 5, No. 5, 1984

Complement:continuedfiom p. 121 factor B. Both C2 and factor B require to be complexed with their partners, C4b and C3b before they can be efficiently cleaved by the pre-existing enzymes C 1~ and factor D. Now that several of the complement proteins have been cloned, as was reported at the meeting, we may expect some of the three-dimensional questions to be answered indirectly by site specific mutagenesis experiments. There is now good agreement on the outline of the mechanism by which the triggered enzymes of the complement system lead to cell lysis. Both S. Bakdi (Giessen) and H. M/.iller-Eberhard (La Jolla) produced beautiful electron micrographs of the membrane attack complex (p. 124). This is composed of one molecule each of C5, C6, C7 and C8 and a variable number (8-18) of C9 molecules. While the C5-8 molecules attach to the cell surface, C9 inserts itself into the membrane bilayer and polymerizes into a ring structure, leaving a central channel through which ions and other small molecules can pass. Poly-C9 can be produced by in-vitro manipulation of the purified component, and this will insert into pure lipid liposomes, though not into cells, probably because of electrostatic repulsion. It requires but a few of these unscalable punctures to lyse a nonnucleated cell. The formation of the membrane attack complex is a remarkable process: water-soluble proteins develop an affinity for surface structures and one of their number undergoes a polymerization process in the course of which it develops a strongly hydrophobic property. The proper study of complement genes has begun with the cloning of C3 (G. Fey, La Jolla), C4 (M. Carrol, Oxford), C2 and factor B (R. Campbell and D. Bently, Oxford) and, most recently, Clq (K. Reid, Oxford). The latter work is at a preliminary stage, with interesting results anticipated from the study of families in which normal C l q appears to be replaced by an abnormal protein. Though sequence analysis reveals that there is an abnormal structure with a kink in the collagenous part of the molecule, it remains possible that this abnormal molecule is present in normal subjects, albeit at a much lower level of expression than conventional C 1q. In contrast, investigation of the C3 gene is at an advanced stage in mouse and man. C3 is coded by a single gene in both species and the sequences show strong homologies. Significant homologies are also found with another protein containing a thioester bond, a2-macroglobulin. The scene is set here Continuedon p. 124

124

Continuedfrom p. 122

Immunology Today, vol. 5, No. 5, 1984

125

Immunology Today, vol. 5, No. 5, 1984

Fig. 1. A : Negatively stained, complement-lysed erythrocytes. C5b-9(m) complexes are seen as numerous circular ('classical') lesions over the membrane together with some 'twinned' forms (bold arrows). The C5b-9(m) complexes are seen as 10 nm high, cylindrical projections along the bent edge of the ghost membrane at the top (arrows). The light rim representing the sharply bent membrane in tangential view is attenuated or interrupted at the site of attachment of the complexes. B : Negatively stained preparation of isolated CSb-9(m) complexes in detergent solution. The complex has the basic structure of a 15 nm high, thin-walled cylinder, rimmed by an annulus at one end. The cylinder is seen in various levels of tilt between side views (s) and axial projection (e). C : Selection of CSb-9(m) complexes with a small appendage. Such appendages (arrows) are often seen on the annulus, particularly by low electron dose image recording. This stalk carries antigenic determinants of C5 and C6 (Tschopp et al., 1982 Proc. NatlAcad. Sci. USA, 79, 7474-7478). D : 'Poly-Cg' This is formed by prolonged incubation of purified human C9 in detergent-free buffer solution at 37°C as described by Tschopp et al. (1982) Nature, (London), 298, 534-537. The C9 oligomers exhibit a cylindrical structure closely resembling the C5b-9(m) complex except for the absence of appendages on the annulus. Occasionally, small ordered arrays of cylinders are seen (arrows), associated at the putative apolar terminus opposite the annulus. E and F : CSb-9(m) complexes (arrows) generated on erythrocytes, purified and re-incorporated into phosphatidylcholine liposomes. Vesicles that escaped incorporation of a complex (asterisks) are characteristically impermeable to the stain. Typically, the complexes project 10 nm exterior to the plane of the membrane. G : Complementary freeze-etch replicas of antibody-sensitized sheep erythrocytes, lysed with human complement. Fracture E-faces (left frame) exhibit numerous ring-shaped structures, interpreted to represent the intramembranous portion of C5b-9(m) cylindrical complexes. The rings are complementary to circular defects in the lipid plateau of the inner membrane leaflet (PF-face). A number of complementary lesions are labelled by arrowheads. Inset at upper right shows C5b-9(m) annuli (arrows) on the etched true outer surface (ES) of a proteolytically stripped ghost membrane. 25 ° rotary shadowing with Pt. Scale bars indicate 100 nm in frames A to G. Sodium silicotungstate was used for negative staining in frames A to F. Micrographs kindly supplied by S. Bhakdi andJ. Tranum-Jensen (see Biochim. Biophys. Acta," 1983, 737,343). for investigations of gene expression and alteration of the sequence expressed so that structure/function relationships can be more clearly understood. It has been known for some years that C2, factor B (By')and C4 are coded within the major histocompatibility complex ( M H C ) . T h e C4 gene is duplicated in both mice and men, and in the case of mice, one o f the duplicates is not haemolytically active. However, the control of expression of the two genes is very complex (D. Shreffler, St. Louis), being variably testosterone-dependent in different strains and subject to control from genes not in the M H C . The idea that the C4 gene is merely duplicated is an oversimplification. O n e copy of the gene (Slp) codes for the non-active variant, which has the thiolester site in a different position from C4 and cannot be cleaved by C1. In some h u m a n s there are chromosomes which code for three gene products, and overlapping cosmids have been found for one such array, interestingly, not all in the same transcriptional orientation (Carrol). In some of the more unusual types of C4 expression in mice, as many as six C4 genes have been found (R. Tosi, Paris). Complete maps of the M H C - l i n k e d genes have been made for both m a n (Carrol, Campbell and colleagues) and mice (D. Chaplin and colleagues, Boston). In both cases the order is 5 ~- C 2 - B f - - - C 4 - 3 . The Slp gene of mice is closer to Bfthan true C4, as is C4A than C4B in humans. The C4A gene has an additional 7kb insert in one of the introns. The C2 and Bfgenes are very close and separated from the C4 genes by about 50 kb. Detailed study of the B f gene reveals 18 exons, three of which

represent internal duplications close to the 5' end. Work on the C2 gene is at a much more preliminary stage: the 5' end of the gene has not been located and the details of its homology with the Bar gene are not yet known. It will be particularly interesting to see if the internal repeat is present in the C2 gene and the Bfgene, as this may give some indication of their history, a topic which was the subject of interesting speculation by W. Bodmer (London). The control of the expression of the complement genes, especially those coded within the M H C was discussed by H. Colten (Boston) who showed that levels of m R N A are subject to different control in the hepatocyte and the monocyte. W h e n the latter is maintained in culture, it matures into a macrophage and this is associated with greatly increased secretion of C2 and factor B synthesis, due to increased messenger production. Not all macrophages are equal, however, as the ratio of C2 and factor B differs between alveolar and breast milk macrophages. C4 expression is in some cases subject to a specific feedback control dependent on the C4 concentration of the culture supernatant. The contributions of J. Edwards (Oxford) and P. Lachmann (Cambridge) were cautionary in tone. Edwards pointed out that there is now a crisis of understanding in the genetics of a system as complicated as C4, in which the n u m b e r of identifiable haplotypes, defined by protein and D N A polymorphism, exceeds the n u m b e r which is expected by conventional theories of generation and fixation of alleles. It will be interesting to observe how this brain

pain will be resolved: will data or thought have the dominant role? L a c h m a n n reviewed the current state of knowledge on complement deficiencies in m a n and emphasized that it is possible, when considering the pathogenesis of many diseases, to distinguish between the involvement of a deficiency itself and that of a linked gene. This is especially true of the early complement components in which primary genetic deficiencies and deficiency due to secondary loss of the complement activity are often associated with the same diseases. It is probable that the increased susceptibility to severe neisserial infections of those with terminal pathway deficiency is fhnctional in origin. H e also reviewed in detail the study by Yukijamaet al. on C9 deficiency in the Japanese (to be published in the proceedings of the 5th International Congress of Immunology). This seems to be a significantly c o m m o n defect in the Japanese population (based on investigations of close on 105 individuals), but that there is no evidence to suggest that this condition is associated with increased disease susceptibility. It seems fair to conclude that complement research has now reached the point of maturity where inside information on nomenclature and experimental reliability are no longer necessary to the understanding of the subject. The complement pathways may therefore provide interesting models for those interested in the biology, pathology and genetics of complex systems.

Michael Hobart is in the Mechanisms in Turnout Immunity Unit, MRC Centre, Cambridge, UK. ~,~:.

-

Agreement now the norm.

Many immunologists hold that complement is baffling or irrelevant or, most conveniently, both but a recent meeting(∗) emphasized that complement is in...
4MB Sizes 2 Downloads 3 Views