and Immunity (2006) 7, 439–449 & 2006 Nature Publishing Group All rights reserved 1466-4879/06 $30.00 www.nature.com/gene

REVIEW The genetics of the amyloidoses: interactions with immunity and

J Buxbaum Department of Molecular and Experimental Medicine, The Scripps Research Institute, La Jolla, CA, USA

Historically, the amyloidoses have been associated with inflammation and the immune response. From Virchow’s original description in human pathologic inflammatory states through their identification in horses used to produce antitoxin to their frequent occurrence in the course of multiple myeloma and a somewhat abortive designation as ‘gammaloid’, the disorders were felt to have an inflammatory origin. These presumptive associations antedated the availability of a reliable method for tissue extraction that would allow chemical analysis of the major deposited molecules. With the identification of the multiple precursors and the realization that most were not intrinsic elements of immune/inflammatory pathways, the investigative emphasis shifted to the analysis of the biophysical events involved in aggregation and fibril formation. As more in vivo models and better tools for examination of tissues have become available, it appears as if inflammation may participate as both a response to, and an amplifier of, the effects of the fibrillar aggregates. Hence, while only a limited number of precursors are involved in immunity and inflammation per se, host defense, in its broadest sense, is likely to be involved in the clinically relevant amyloidoses. Further it now appears that harnessing the immune respone in an appropriate fashion may be able to play a role in treatment. Genes and Immunity (2006) 7, 439–449. doi:10.1038/sj..6364323; published online 13 July 2006

Definition A protein preamble Amyloidosis1,2 is a pathologic diagnosis made on biopsy The relationship between in vitro amyloid formation or autopsy when fixed tissues stained with hematoxylin and the amyloidoses: and eosin display extracellular, homogeneous eosinophi- In recent years, recombinant technology has allowed lic regions typically, but not exclusively, in kidneys, the bacterial synthesis and purification of a large number , peripheral nerve, and spleen, depending on of , including those known to form amyloid in the protein precursor. The same areas bind the dye vivo. This has been extremely useful in allowing the Congo red displaying green bi-refringence when exam- experimental dissection of the processes of aggregation ined under polarized light. The deposits are also reactive and fibril formation that are intrinsic to the pathogenesis with specific for serum amyloid P-compo- of these disorders. In addition, other disease-related nent,3 E4 and the heparan sulfate proteo- proteins, including a-synuclein, the major component of glycan perlecan.5 With the electron microscope, the Lewy bodies in Parkinson’s disease and Huntingtin the deposits contain discrete nonbranching fibrils with aggregated poly-glu protein found in the cellular diameters ranging between 8 and 10 nm.6 Extraction inclusions in Huntingtin’s disease, when incubated and purification of the Congophilic fibrillar components under amyloid-forming conditions in the test tube, form from such tissues have revealed a cross-b structure.7–9 fibrils with similar biophysical properties as the proteins involved in the human amyloidoses.10,11 In vivo these disorders do not have the histopathologic characteristics of the human amyloidoses and technically should not be referred to as such; however, it is possible that the same Introduction biophysical principles will prevail for these proteins as Twenty-six different amyloid precursor proteins have for those forming amyloid in vivo. been identified in various human amyloidoses and the The proteins that have in vivo disease potential are disorders have now been classified according to the inherently small (o30 kDa) or are cleavage products of nature of the precursor, rather than the less precise larger molecules, suggesting that the amyloidogenic clinical nomenclature utilized in the past (Table 1). potential of a domain is reduced by its structural context, that is, embedded in a larger molecule, which requires either cleavage or denaturation to allow aggregation to Correspondence: Dr J Buxbaum, Department of Molecular and begin. Some of the precursors seem to achieve the same Experimental Medicine, The Scripps Research Institute, 10550, N. protection by existing as a complex with other proteins, Torrey Pines Road, MEM 230A, La Jolla, CA 92037, USA. E-mail: [email protected] for example, immunoglobulin light chains with them- Received 25 May 2006; accepted 30 May 2006; published online 13 selves as homodimers or in the H2L2 tetramer or b2- July 2006 microglobulin noncovalently complexed on the cell The genetics of the amyloidoses J Buxbaum 440 Table 1 The human amyloidoses9

Amyloid Precursor protein Systemic (S) or localized, organ restricted (L) Syndrome or involved tissues

AL Immunoglobulin light chain S, L Primary; myeloma-associated AH Immunoglobulin heavy chain S, L Primary; myeloma-associated

Ab2M b2-microglobulin S Hemodialysis-associated L?, Joints ATTR S Familial senile systemic L? Tenosynovium AA (Apo)serum AA S Secondary, reactive AApoAI Apolipoprotein AI S Familial Aorta AApoAII Apolipoprotein AII S Familial AApoAIV Apolipoprotein AIV S Sporadic, associated with aging AGel Gelsolin S Familial (Finnish) ALys Lysozyme S Familial AFib a-chain S Familial ACys S Familial ABri ABriPP S Familial dementia, British ADana ADanPP L Familial dementia, Danish Ab Ab protein precursor (AbPP) L Alzheimer’s disease, aging APrP protein L Spongioform encephalopathies ACal (Pro) L C-cell tumors AIAPP Islet amyloid polypeptide L Islets of Langerhans AANF Atrial natriuretic factor L Cardiac atria APro L Aging pituitary AIns Insulin L Iatrogenic AMed Lactadherin L Senile aortic, media AKer Kerato-epithelin L Cornea, familial ALac Lactoferrin L Cornea A(tbn)btbn L Odontogenic tumors

Proteins are, when possible, listed according to relationship. Thus, are grouped together, as are polypeptide hormones. aADan comes from the same gene as ABri. bA(tbn), to be named. The designation is waiting for a protein name.

surface with major histocompatibility complex (MHC) amyloid fibrils deposited in tissue parenchyma. Recent class I heavy chains. Even more interesting are the studies with some precursors in tissue culture systems observations that fragments or domains of some pro- have suggested that small oligomeric aggregates that are teins, not shown to be associated with disease or in vivo preprotofibrillar may be toxic to cells.16–18 Definitive deposition, when incubated under the appropriate evidence that this occurs in vivo has not yet been conditions of concentration, ionic strength and pH will presented, although there are experiments suggesting also behave in a fashion similar to those proteins that this might be the case.19 responsible for the clinical amyloidoses.12 Hence, the notion has arisen that the capacity to form the cross-b- structure characteristic of the amyloid protofibril is an Mechanisms of amyloidogenesis in intrinsic property of many proteins; hence, must have some structural features that have been selected through humans evolution as useful, particularly in the context of multi- There are several general processes that lead to clinical domain molecules. Several examples have now been (Figure 1). In one group of precursors, described in which the amyloid structure serves an overproduction or failure to excrete an intact molecule apparently normal physiologic function, notably the curli that has a concentration dependent tendency to misfold,

protein of escherichia coli, yeast and the melano- for example, ANF, b2 microglobulin and IAPP, is some melanin binding protein Pmel17.13–15 sufficient.20–22 These molecules require no other proces- Our discussions in this review will cover only those sing to form fibrils, although partial might conditions in which tissue deposits fulfilling the patho- enhance their fibril forming capacity. In a second class, logic criteria for amyloid, that is, Congophilia, ultra- the intact molecule is not amyloidogenic but abnormal structurally fibrillar, co-deposited SAP, ApoE and site specific proteolysis generates an amyloidogenic Perlecan, are found. For the most part the deposits are fragment, e.g Alzheimers Disease (Ab).23 In a third extracellular, although there are now instances in which pathway, inadequate normal proteolytic cleavage yields it has been reported that the earliest steps in aggregation an amyloidogenic fragment. This may occur when the can be seen intracellularly. substrate is overproduced and there is a relative For most of the history of amyloid research, the insufficiency of the normal degradative pathway with manifestations of the disease were felt to be the result respect to the increased precursor load, as may be the of tissue replacement or displacement by the masses of case in AA and perhaps in associated amyloidesis (AL),

Genes and Immunity The genetics of the amyloidoses J Buxbaum 441 Amyloidogenesis

CATABOLISM Familial Amyloid Polyneuropathy Alzheimer’s Reactive Spongiform disease Amyloidosis encephalopathies

TTR A PP SAA Prion

A AA PrPSC

?

ß-Sheet Protein X Pre-Fibrils

Perlecan ProtoFibrils P-Component ApoE Other factors ?

Congo Red +, SAP +, ApoE + Fibrils Figure 1 Four general pathways of amyloidogenesis are shown. In transthyretin FAP, the fibril precursor is a misfolded monomer formed after native monomer is released from the tetramer. In Alzheimer’s disease the fibril precursor, Ab, is released by a series of cleavages from the precursor AbPP (Ab protein precursor). In reactive (AA) amyloidosis, the production of the SAA precursor exceeds the capacity of the normal degradative mechanisms to completely digest the core AA molecule. This is also likely to be true in AL. In the prion diseases, no processing of the native protein appears to be necessary in the face of adequate amounts of misfolded template. that is, a combination of overproduction and partial deposits in which the fibrils appear to be derived from cleavage.24,25 In this instance, there are few experimental cytokeratin.30,31 It has not been established that the data available regarding the details of these processes. cytokeratins are of wild-type sequence. Further the In the case of the homotetrameric protein transthyretin process of fibril formation has not been examined (TTR), a shift in the equilibrium between tetramer and experimentally. It is not clear why some of these patients monomer leads to the enhanced population of a have localized procalcitonin or cutaneous amyloid and monomer pool predisposed to misfold. The equilibrium others do not. The distinction may totally reside in the is shifted toward the misfolded monomer by the amount of precursor that is available. Alternatively it presence of any of 80 þ of mutations.26 It is also likely may result from the action of alleles at loci that impact on that proteolysis will accelerate the process. In the case of the process of amyloidogenesis or influence the removal the prionoses, the native proteins appear to require no of protein aggregates from local areas. In formal genetic particular primary sequence change or cleavage and may terms, this would be assessed by determining interaction not be overproduced. It seems that they only require the effects between the major gene responsible for the presence of an amyloidogenic folding template, that is, primary disease and potential candidate genes hypothe- the prion seed.27 sized to be involved or family studies in which the property segregated in a defined fashion. Additional instances in which a primary genetic Amyloidosis as a consequence of disorder leads to amyloidosis are found in the hereditary disorders of the innate inflammatory pathway, including a primary genetic disorder Familial Mediterranean Fever (FMF), etc (see below). The hereditary aspects of this subset of amyloidoses These are inherited in either an autosomal recessive reside in conditions that increase the production of a or dominant fashion and result in recurrent episodes of protein that at normal concentrations does not form systemic inflammation and increased production of significant amounts of amyloid. In the genetic disorder SAA, the Amyloid A (AA) precursor.32 Individuals, multiple endocrine neoplasia (MEN) type 2A indivi- bearing particular alleles of SAA genes, are at increased duals, who have mutations in the RET proto-oncogene risk for hepatic and renal amyloidosis (see below and represent 20% of cases of medullary carcinoma of the Table 5). In these cases, the amyloid precursor has the thyroid (MCT) in which the overproduction of procalci- wild-type structure, the mutation being in the molecule tonin leads to local amyloid formation.28 Histologically responsible for the exuberant inflammatory response. demonstrable procalcitonin-derived amyloid is a major The other class of amyloidoses that has a major genetic feature in 30–40% of MCT, the majority of which are not component includes autosomal dominant disorders in associated with MEN 2A.29 Hence, the genetic lesion which a structural abnormality in a potential amyloid predisposes to the tumor, not its tendency to form a substrate, usually a mutation, generates a protein with a particular species of amyloid. One-third of MEN 2A pathologic gain of function, that is, the property of patients, apparently all with a mutation at position 634 in becoming insoluble under physiologic conditions. These RET, have cutaneous lichen amyloidosis, Congophilic include the disorders related to mutations in transthyretin,

Genes and Immunity The genetics of the amyloidoses J Buxbaum 442 fibrinogen, gelsolin, the Abeta precursor protein (APP), b2-Microglobulin (dialysis-AL) 33–39 , lysozyme and cystatin. Consider- Amyloid deposition consisting of b2-microglobulin, the able progress has been made in understanding the light chain of class I MHC molecules expressed on the process of in vitro fibrillogenesis in almost all of these cell surface is very common among patients with chronic cases; however, the same precision has not been achieved renal failure, who are maintained on chronic hemodia- in studying the processes in vivo. lysis, being virtually universal after 15 years of such Mutations in genes that encode molecules that process treatment.53 There does not appear to be any genetic a potentially amyloidogenic substrate into a nonamyloi- component determining either the frequency or time of dogenic can also result in clinical amyloidosis. appearance of the deposits, which is a function of both The most common mutations in early onset, autosomal the nature of the dialysis membrane and characteristics dominant Alzheimers Disease are in the genes of the dialyzing fluid.54,55 Attention to both has resulted PS-1 and PS-2, which are components of the a-secretase in a gradual decline of this iatrogenic form of amyloi- complex, and result in alterations in the cleavage dosis. patterns of AbPP with increased concentrations of the 40–42 amyloidogenic Ab1À40 and Ab1À42 precursors. It is Amyloid A possible that polymorphisms in genes that process The third form of systemic amyloidosis of intrinsic transthyretin, anf, iapp, apoA1, prolactin and others that interest to immunologists is that produced by the process or chaperone Ab may contribute to the deposi- deposition of AA fibrils.56 AA is a cleavage product of tion of these wild-type molecules late in life, although at the normal serum protein SAA, which circulates in a this time there is little genetic evidence to suggest that complex with high-density and is highly this is the case. elevated in the course of acute and chronic infectious or noninfectious inflammatory responses.57,58 The protein appears to play a role in handling by macrophages, and there has been some speculation that this process Amyloidoses derived from molecules may be actively involved in atherogenesis.59 The hepatic involved in immune function transcription of the gene is highly sensitive to interteukin (IL)-6, which in turn may be influenced by IL-1 and The immunoglobulin associated amyloidoses alpha (TNFa). The gene is also The amyloidoses that are most intimately related to the actively transcribed in cells other than hepatocytes.60 immune system are those derived from immunoglobulin Historically, prior to the availability of relatively (Ig) chains. AL, the amyloid consisting of immunoglo- effective anti-inflammatory therapy, as many as 25% of bulin light chains or their fragments and the far less patients with long-standing (RA), a commonly seen AH, in which the amyloidgenic heavy classic autoimmune process, developed renal, hepatic chain fragment results from a domain length deletion in and splenic amyloidosis, with renal failure a common the Heavy chain gene, are both secreted products of cause of death.61 With more aggressive modes of therapy monoclonally expanded B cell populations. The L-chains the incidence is now approximately 5% in the US with may be intact or truncated while the H-chain related only half of those with positive screening biopsies proteins usually have internal deletions.43 The prolifera- (abdominal fat or rectal) having clinically significant tive disorder may have the clinical neoplastic character- deposition.62 istics of multiple myeloma or those of a monoclonal A prolonged inflammatory state is required for AA gammopathy of undetermined significance (MGUS).44 In deposition but it is not sufficient. The amount of SAA either case, the protein product is pathogenic producing precursor available plays some role since it has been renal, cardiac or neuropathic clinical symptoms when it shown that keeping the serum SAA level below 10 mg/l deposits in those organs. The somatic cell genetics of prevents clinical progression; however, it is unclear that these disorders have been extensively studied from the the risk for amyloid deposition increases once one goes point of view of oncogenesis as have the structural above that level.63 It appears that the propensity toward features of kappa and lambda light chains, which appear amyloid deposition in the presence of systemic inflam- to favor amyloid formation and will not be discussed mation is determined by its severity and chronicity, and here.45–47 conditioned by specific alleles encoding the various There are several reports of the familial incidence of isoforms of SAA. The relevant susceptibility alleles vary myeloma, the germline genetics of which have not been in different populations (Table 3). established.48 There are a smaller number of reported The three SAA isoforms (SAA1, 2 and 4) are encoded instances of familial AL but no genetic studies have been by three homologous genes (Table 2).64,65 All are located performed on any of these very rare kindreds49,50 on human 11 (and mouse chromosome 7).66 Interestingly, a number of studies have reported that One, SAA4, encodes a molecule which is 52% homo- African-Americans have a higher frequency of multiple logous at the amino-acid level and has not been found to myeloma and examination of the Department of Veterans be an amyloid component. There is also a pseudogene Affairs inpatient data base suggests that they also have a which is not expressed. The SAA1 gene has five alleles higher frequency of AL.51 Similarly, an extensive autopsy (Table 2). SAA2 has two. Both SAA1 and 2 proteins have series indicated that Hispanics of Mexican descent been identified as amyloid components in humans while seemed to have a higher prevalence of both AL and SAA2 is the major fibril component in mice. AA, but neither observation has been confirmed by a Over the last 10 years, it has become clear that the formal epidemiologic study nor have they been subjected distribution of SAA alleles in individuals who develop to genetic analysis; hence, the possibility remains that the amyloid in the course of inflammatory diseases differs effect may be environmental rather than genetic.52 from appropriate population controls. Further it appears

Genes and Immunity The genetics of the amyloidoses J Buxbaum 443 Table 2 Human SAA genes and proteins64

SAA1.1 V52A A57V G73D SAA1.2 V52A A57V SAA1.3 V52A A57V D60N SAA1.4 A57V SAA1.5 V52A A57V SAA2.1 V52A A57V D60N F69L G70T N84K G90R SAA2.2 V52A A57V D60N F69L H72R N84K G90R

Amino-acid sequences are identical except for positions noted: A, alanine; D, aspartic acid; F, phenylalanine; G, glycine; H, histidine; K, lysine; L, leucine; N, asparagines; R, arginine; T, threonine; V, valine.

Table 3 (SAA) alleles in rheumatoid arthritis

Population Diagnosis SAA 1 genotype no. (%) SAA 1 allele frequency

1.1 1.2 1.3 2.2 2.3 3.3 1 2 3

Japanese RA+, AA+ 1 (6.7) 0 0 2 (13.1) 3 (20) 9 (60) 2 7 21 Japanese RAÀ,AAÀ 4 (8) 14 (28) 9 (18) 4 (8) 10 (20) 9 (18) 31 32 37 Japanese RA+, AA+ 2 (4) 3 (6) 8 (16) 6 (12) 17 (34) 14 (28) 15 32 53 Japanese RA+, AAÀ 6 (12) 10 (20) 10 (20) 3 (6) 12 (24) 9 (18) 32 28 40 Japanese RA+, AA+ 0 (0) 5 (11.6) 5 (11.6) 3 (7) 15 (35) 15 (35) 10 26 50 Japanese RA+, AAÀ 6 (7.8) 13 (16.9) 26 (33.8) 12 (15.6) 14 (18.2) 6 (7.8) 51 51 52 Japanese RAÀ,AAÀ 12 (11.7) 18 (17.5) 25 (24.3) 9 (8.7) 26 (25.2) 13 (12.6) 67 62 77 Japanese RA+,AA+ 18 54 34 Japanese RA+, AAÀ 34 35 25 US (Cauc) RA+, AA+ 29 0 3 US (Cauc) RAÀ,AAÀ 101 4 35 UK RA+, AA+ 22 (85) 4 (15) 0000 4840 UK RA+, AAÀ 18 (69) 5 (19) 0 3 (12) 0 0 41 11 0 UK RAÀ,AAÀ 55 (58) 27 (28) 7 (7) 3 (3) 3 (3) 0 144 36 10 Korean RAÀ,AAÀ 7 (7.8) 19 (21.1) 21 (23) 7 (7.8) 17 (18.8) 18 (20) 54 50 74 Taiwanese RAÀ,AAÀ 10 (10.5) 17 (17.9) 13 (13.7) 15 (15.8) 24 (25.3) 15 (15.8) 50 71 67

that particular alleles may impart increased susceptibil- Recent findings have suggested that another ity in some populations but not in others, and in at least polymorphism in the SAA 1 gene, at position À13 in one instance interactions of alleles at different snp’s in the 50 regulatory region of the gene is more closely the same gene appear to increase the risk. associated with the occurrence of amyloid in both In Asian (Japanese, Korean, Taiwanese) control popu- Japanese and Caucasian RA patients than any of the lations, the SAA 1 allele frequencies are quite similar prior exon 3 based allelic associations.69 The frequency of (Table 3).67 The three alleles (1.1, 1.3 and 1.5) being the TT allele is higher in Japanese than in the Caucasians. relatively equally represented with some skewing to- The association of the TT and SAA 1.3 alleles is not ward SAA 1.3. In Caucasian groups, SAA 1.1 is much significantly greater in the amyloid patients (P ¼ 0.09), more common being significantly higher than in the but a larger sample size might demonstrate a synergistic Japanese controls.68,69 In Japanese RA patients without a effect of the two snp’s. A recent functional study tissue diagnosis of amyloidosis, the allele distributions suggests that the TT allele has greater transcriptional do not differ from controls without RA (Table 3).70 activity, a finding that would predict an interaction However, in several studies of Japanese patients with between them.73 RA and amyloid SAA 1.3 was significantly over- Two studies of Japanese RA patients with and represented, particularly in the homozygous state.71,72 without amyloidosis did not show differences in the One analysis, rather than showing a significant increase distribution allele frequencies at the SAA 2 locus.67,74 in SAA 1.3, showed only a decrease in the 1.1 allele in the However, an analysis of juvenile chronic arthritis (JCA) amyloid patients (P ¼ 0.002).69 In that same study, the patients in the UK showed that the a-allele was more increase in SAA 1.3 was not quite significant (P ¼ 0.06), common in those with amyloid than in those without suggesting perhaps that the study did not have sufficient amyloidosis, but that allele was also more common in power. The difference between the RA patients with both JCA groups than in the controls (P ¼ 0.04), suggest- amyloidosis and the population controls without RA was ing perhaps that the allele may play a role in the statistically more impressive (P ¼ 0.01). In US patients pathogenesis of JCA.75 Among patients with JCA in with RA the frequency of SAA 1.1 is significantly higher Great Britain, SAA 1.1 is significantly increased in those in patients with amyloid than in nonrheumatoids with- with amyloidosis over both JCA patients without out amyloid, not unexpectedly resembling the UK amyloid and controls; however, the difference in the distribution more than that in Japan.69 frequency of SAA 1.1 between adult RA patients from the

Genes and Immunity The genetics of the amyloidoses J Buxbaum 444 Table 4 Serum amyloid A (SAA) alleles in juvenile chronic arthritis

Population Diagnosis SAA 1 Genotype no. (%) SAA 1 allele frequency

1.1 1.2 1.3 2.2 2.3 3.3 1 2 3

UK JCA+, AA+ 11 (50) 7 (31.8) 1 (4.5) 2 (9.1) 0 1 (4.5) 30 11 3 UK JCA+, AAÀ 10 (37) 1 (3.7) 9 (33) 0 2 (7.4) 5 (18.5) 30 3 21 UK JCA+, AA+ 33 (80) 7 (17.1) 1 (2.4) 0 0 0 74 7 1 UK JCA+, AAÀ 1 (12.5) 6 (75) 1 (12.5) 0 0 0 9 6 1

Table 5 Serum amyloid A (SAA) alleles in familial mediterranean fever

Population Diagnosis SAA 1 Genotype no. (%) SAA 1 Allele frequency

1.1 1.2 1.3 2.2 2.3 3.3 1 2 3

Turkish FMF+, AA+ 54 (74) 16 (22) 1 (1.4) 1 (1.4) 1 (1.4) 0 125 19 2 Turkish FMF+, AAÀ 27 (27) 42 (42) 3 (3) 22 (22) 5 (5) 1 (1) 99 87 10 Turkish FMFÀ,AAÀ 19 (19) 43 (43) 4 (4) 25 (25) 9 (9) 0 85 92 13 Armenian FMF+, AA+ 22 (46.8) 12 2 9 1 1 58 25 5 Armenian FMF+, AAÀ 15 (16.7) 38 1 34 1 0 69 107 2 Israeli FMF+, AA+ 22 (35.5) (20%) (2.5%) (37.5%) (2.5%) Israeli FMF+, AAÀ 34 (15.8) (35%)

same center with and without amyloid as a complication would be responsible for the occurrence of amyloidosis was not significant (P ¼ 0.07) (Table 4).68 in this disorder.84 Similar analyses have been performed in populations The comparison of the SAA allele distributions in these defined by the mutations in genes that are responsible disorders with those in RA is interesting because the two for the hereditary inflammatory diseases FMF, Familial sets of disorders have arisen in populations with genetic Cold Urticaria (FCAS), Muckle–Wells Syndrome (MWS), histories that are likely to be quite different. Studies of Hyperimmunoglobulinemia D with periodic fever syn- FMF kindreds indicate that one must be homozygous for drome (HIDS), TNF-receptor associated periodic syn- an FMF pyrin gene mutation (most commonly M694V) to dromes (TRAPS), Neonatal-onset inflammatory disease/ display the febrile phenotype. However, the appropriate chronic infantile neurologic cutaneous and articular allele of the SAA1 precursor isoform increases the risk of syndrome (NOMID/CINCA) and Pyogenic arthritis with developing amyloid deposition in the course of the pyoderma gangrenosum and acne (PAPA).76–79 All of periodic inflammation, particularly if one is homozygous these are characterized by recurrent febrile episodes with (Table 5). It appears that, in the presence of the same various constellations of clinical features. AA amyloido- pyrin mutation (M694V), homozygous SAA1.1 Israeli, sis is common in FMF, rare in HIDS, uncommon in Armenian and Turkish FMF patients are at greater risk FCAS, present in about 10% of cases of TRAPS, and 25% for amyloidosis.85,86 These observations reinforced, but of MWS. No cases have yet been reported in PAPA and did not fully conform to, findings in RA patients (see only a few in NOMID/CINCA cases who reach adult- above). In both the Israelis and Armenians, male sex as hood. As in sporadic AA deposition is largely renal, well as the SAA1.1 allele are associated with a higher hepatic and splenic. We will not discuss the genetics of frequency of renal amyloidosis.87,88 these disorders since several recent reviews have covered A similar analysis in patients with the HIDS, who the subject in great detail. All have mutations in genes rarely have amyloid, did not show any difference in the involved in apoptotic events in mediated distribution of SAA alleles compared to population inflammatory pathways except for HIDS where the controls despite the fact that they had very high SAA mutations are in mevalonate gene. The relation- levels. No studies were carried out to determine the ship of this gene to intrinsic inflammatory processes is distribution of the À13 T/C alleles in the SAA unclear.77,78,80–83 promoter.89 The data suggest that prolonged SAA The prototypic and most frequently described of these elevations alone may not be sufficient to uniformly disorders, FMF is autosomal recessive as is HIDS. The result in AA deposition, leaving the way open for the others appear to have dominant inheritance. The action of other gene products to play a role. relatively large number of individuals with FMF that have been studied in detail, and the fact that the disorder has occurred on a variety of ethnic Interfaces between the amyloidoses and backgrounds has allowed some insight into the patho- genesis of the disease. On clinical grounds, it was the immune system predicted more than 10 years ago that another gene in The pathologic examination of amyloid infiltrated tissues addition to that responsible for the FMF syndrome has generally failed to reveal any of the conventional

Genes and Immunity The genetics of the amyloidoses J Buxbaum 445 hallmarks of inflammation. In some experimental mod- The success of this approach has led to similar animal els of AA, however, there appears to be considerable studies in models of AL and the transthyretin amyloi- mononuclear activation. Some have interpreted this as doses with some success.94,105 These will presumably reflecting an attempt to clear the deposits.90 The lead to human trials with appropriately humanized alternative explanation is that the macrophages contri- monoclonals that have been effective in the murine bute to deposition by cleaving the precursor SAA, to models. release AA capable of forming fibrils in situ. Mixed cultures of SAA-producing hepatoma cells or hepato- cytes and macrophages have been reported to produce Other genetic lessons learned from the AA fibrils in tissue culture.25 A single published mixed culture experiment with human bone marrow cells and amyloidoses macrophages from an amyloid patient suggested a The interaction of specific alleles at the SAA loci with the similar process.91 More recent studies in which L-chains genes encoding the hereditary periodic inflammatory isolated from patients with AL and light chain deposi- disorders to predispose to amyloid formation is an tion disease were incubated with cultured human instructive example of the emerging concept that there mesangial cells seem to demonstrate cleavage of the are no single gene diseases. Another, perhaps more amyloidogenic L-chains with subsequent fibril forma- subtle, instance is seen in patients with Familial tion, a very interesting result awaiting confirmation in Amyloidotic Polyneuropathy (FAP) related to mutations other laboratories.92 in transthyretin. The homotetramer is composed of four In vivo models in which L-chain fibrils are injected 127 amino-acid monomers. Over 100 mutations have subcutaneously in mice have shown resorption of the been reported at 50 positions in the peptide. Ninety of aggregates with the major cellular player appearing to be these appear to be associated with tissue amyloid neutrophil. The process then can be accelerated by deposition. (http://www.bumc.bu.edu/Dept/Content. passive administration of an apparently conformation aspx?departmentid ¼ 354&PageID ¼ 5530). specific .93,94 While many of the mutations have only been reported In contrast to the systemic amyloidoses, it has been in a single or small number of kindreds, it has been well recognized that glial activation is common, if not suggested that particular mutants may be associated universal, in the neighborhood of Ab plaques in the with characteristic clinical presentations,106 in a fashion Alzheimers Disease .95 It has been experimentally similar to the case in cystic fibrosis where the severity of shown that Ab fibrils can activate glia and this may be the disease varies with the particular mutation. Among mediated by the nuclear factor kappa B pathway the TTR mutants only a few have been seen in enough triggered via RAGE or CD36.96–98 Glial activation seems individuals to derive a statistically valid impression to result in more Ab production as well, again suggesting regarding the association between the mutation and that an apparent defense mechanism may actually clinical presentation. FAP, in which the valine at position enhance the pathologic process. It remains an open 30 is replaced by methionine (TTR V30M), has been question whether these responses are intrinsic to the reported in approximately 1500 individuals in 500 basic process of fibril formation or whether they kindreds in Portugal, as well as in Japan and in represent host defenses to the abnormal conformers, Sweden.107,108 In Sweden, the allele shows low clinical which are inadequate to eliminate the insult but by penetrance and late onset. While the disorder was known responding actually increase the damage. to behave as an autosomal dominant, almost from the The other side of the inflammatory coin with respect time of its original description, with increasing numbers to the amyloidoses is represented by the relatively of carriers analyzed it has become clear that the successful attempts to immunize both experimental phenotype varies with respect to age of onset, the animals and humans to the fibril precursor. The initial presence of cardiac and renal involvement in both attempts to immunize mice transgenic for human Portugal and Japan. A recent analysis of Portuguese mutant APP genes with the amyloidogenic fragment gene carriers has suggested that the age of onset is were startlingly successful, with impressive plaque influenced by interactions among genes known to be resolution.99 The results were significant enough to result involved in either the process of deposition or the carrier in a clinical trial. Unfortunately, perhaps predictably, function of the TTR molecule, that is, ApoE, SAA1, SAA2, about 5% of the immunized individuals developed APCS, RBP4. Single-nucleotide polymorphisms in these severe inflammatory disease of the central nervous genes were examined for their association with early or system.100,101 The frequency of the severe adverse late onset. Those studies did not show significant effects reaction was actually less than one might expect, of any of the five genes analyzed, but demonstrated that considering that while the AD mice were being im- many more interactions among alleles at these loci munized with a human protein (less likely to be occurred in patients with late onset of the disease than homologous to an endogenous functional murine mole- among individuals with early onset.109 These results cule), the trial involved immunizing humans with a suggested that early onset was the default state of the human protein and some autoreactivity was likely. mutation and interactions with the products of some of Nonetheless pathologic examination of the of these alleles delayed the appearance of symptoms, some of the individuals suggested reductions in pla- contributing perhaps 25% to the late onset phenotype. ques.102 Subsequent studies have shown that passive The sample size was not sufficiently large to establish immunization with antibodies directed toward particular whether single gene effects were also present. However, Ab epitopes could be carried out in the transgenic mice, it is worth noting that apart from the alleles at SAA1 and and trials with humanized antibodies of relatively SAA2, which we have described above, prior studies refined specificities are now ongoing.103,104 have shown no association of amyloid deposition with

Genes and Immunity The genetics of the amyloidoses J Buxbaum 446 alleles at APCS or RBP4. There is a well described 11 Harjes P, Wanker EE. The hunt for Huntingtin function: association of ApoE4 with sporadic Alzheimer’s Dis- interaction partners tell many different stories. Trends Biochem ease.4,110–112 However, studies with other amyloidoses Sci 2003; 28: 425–433. including AA, AL and TTR V30M FAP have not shown 12 Chiti F, Webster P, Taddei N, Clark A, Stefani M, Ramponi G any such association. et al. Designing conditions for in vitro formation of amyloid These types of gene interactions could be responsible protofilaments and fibrils. Proc Natl Acad Sci USA 1999; 96: for variation in the phenotype of disorders that in fact are 3590–3594. not related to single major gene effects, for example, any 13 Barnhart MM, Chapman MR. Curli biogenesis and function. Annu Rev Microbiol 2006; 60 (in press). of the autoimmune disorders such as SLE or RA, where it 14 True HL, Lindquist SL. A yeast prion provides a mechanism is already known that multiple alleles of multiple genes for genetic variation and phenotypic diversity. Nature 2000; contribute to the clinical course. 407: 477–483. 15 Fowler DM, Koulov AV, Alory-Jost C, Marks MS, Balch WE, Kelly JW. Functional amyloid formation within mammalian Conclusions tissue. PLoS Biol 2005; 4: e6. 16 Hartley DM, Walsh DM, Ye CP, Diehl T, Vasquez S, The availability of methods to analyze tissue deposits has Vassilev PM et al. Protofibrillar intermediates of amyloid allowed the identification of more than two dozen beta-protein induce acute electrophysiological changes precursor proteins as being responsible for the human and progressive in cortical . J Neurosci and animal tissue pathology we define as amyloidosis. 1999; 19: 8876–8884. While the original, and still the quantitatively major 17 Reixach N, Deechongkit S, Jiang X, Kelly JW, Buxbaum JN. Tissue damage in the amyloidoses: Transthyretin associated disorders, that is, infectious and noninfectious monomers and non-native oligomers are the major cytotoxic inflammation and monoclonal B-cell disorders, have species in tissue culture. Proc Natl Acad Sci USA 2004; 101: their genesis in the immune/inflammatory system, 2817–2822. endocrine, neoplastic, and hereditary conditions are also 18 Magrane J, Smith RC, Walsh K, Querfurth HW. Heat shock etiologically responsible. However, it appears that there protein 70 participates in the neuroprotective response to may be local inflammatory components involved in both intracellularly expressed beta-amyloid in neurons. J Neurosci the pathogenesis of, and the response to, the deposits. 2004; 24: 1700–1706. These may also be under genetic control and contribute 19 Brown WM, Lange EM, Chen H, Zheng SL, Chang B, to the clinical phenotype. The total number and nature of Wiley KE et al. Hereditary prostate in African these genes remains a subject of speculation and active American families: linkage analysis using markers that investigation. map to five candidate susceptibility loci. Br J Cancer 2004; 90: 510–514. 20 Johansson B, Wernstedt C, Westermark P. Atrial natriuretic peptide deposited as atrial amyloid fibrils. Biochem Biophys References Res Commun 1987; 148: 1087–1092. 21 Gejyo F, Yamada T, Odani S, Nakagawa Y, Arakawa M, 1 Virchow R. U¨ ber den Gang der amyloiden degeneration. Kunitomo T et al. A new form of amyloid protein associated Virchows Arch 1855; 8: 364–368. with chronic hemodialysis was identified as B2-microglobu- 2 Osserman E, Takatsuki K, Talal N. Multiple myeloma. The lin. Biochem Biophys Res Commun 1985; 129: 701–706. pathogenesis of ‘amyloidosis’. Studies on the role of 22 Clark A, de Koning EJP, Morris JF. Formation of islet amyloid abnormal gamma globulins and gamma globulin fragments from islet amyloid polypeptide. Biochem Soc Trans 1993; 21: of the Bence Jones (L-polypeptide) type in the pathogenesis of 169–173. ‘primary’ and ‘secondary’ amyloidosis and the ‘amyloidosis’ 23 Hardy J, Selkoe DJ. The amyloid hypothesis of Alzheimer’s associated with plasma cell myeloma. Semin in Hematol 1964; disease: progress and problems on the road to therapeutics. 1: 3–86. Science 2002; 297: 353–356. 3 Pepys MB. Amyloidosis. Annu Rev Med 2005; 57: 223–241. 24 Eulitz M, Breuer M, Eblens A, Linke R. Production of 4 Gallo G, Wisniewski T, Choi-Miura NH, Ghiso J, Frangione B. amyloidogenic from human immunoglobulin light Potential role of apolipoprotein-E in fibrillogenesis. Am J chains. Amyloid Amyloidosis 1990: 505–509. Pathol 1994; 145: 526–530. 25 Kluve-Beckerman B, Manaloor JJ, Liepnieks JJ. A pulse-chase 5 Kisilevsky R. The relation of proteoglycans, serum amyloid P study tracking the conversion of macrophage-endocytosed and apo E to amyloidosis current status, 2000. Amyloid 2000; serum amyloid A into extracellular amyloid. Arthritis Rheum 7: 23–25. 2002; 46: 1905–1913. 6 Cohen AS, Calkins E. Electron microscopic observations on a 26 Foss TR, Wiseman RL, Kelly JW. The pathway by which the fibrous component in amyloid of diverse origins. Nature tetrameric protein transthyretin dissociates. Biochemistry 1959; 183: 1902. 2005; 44: 15525–15533. 7 Glenner GG. Amyloid deposits and amyloidosis. The b- 27 Come JH, Fraser PE, Lansbury Jr PT. A kinetic model for fibrilloses (first of two parts). N Eng J Med 1980; 302: amyloid formation in the prion diseases: importance of 1283–1292. seeding. Proc Natl Acad Sci USA 1993; 90: 5959–5963. 8 Glenner GG. Amyloid deposits and amyloidosis. The b- 28 Saad MF, Ordonez NG, Rashid RK, Guido JJ, Hill CS, Hickey fibrilloses (second of two parts). N Eng J Med 1980; 302: RC et al. Medullary carcinoma of the thyroid: a study of the 1333–1343. clinical features and prognostic factors in 161 patients. 9 Westermark P, Benson MD, Buxbaum JN, Cohen AS, Medicine 1984; 63: 319–342. Frangione B, Ikeda S et al. Amyloid: toward terminology 29 Sletten K, Westermark P. Characterization of molecular forms clarification. Report from the Nomenclature Committee of of calcitonin in amyloid fibrils from medullary carcinoma of the International Society of Amyloidosis. Amyloid 2005; 12: the thyroid. Amyloid Amyloidosis 1990; 113: 477–480. 1–4. 30 Verga U, Fugazzola L, Cambiaghi S, Pritelli C, Alessi E, 10 Conway KA, Harper JD, Lansbury PT. Accelerated in vitro Cortelazzi D et al. Frequent association between MEN 2A and fibril formation by a mutant alpha synuclein linked to early- cutaneous lichen amyloidosis. Clin Endocrinol (Oxf) 2003; 59: onset Parkinson’s Disease. Nat Med 1998; 4: 1318–1320. 156–161.

Genes and Immunity The genetics of the amyloidoses J Buxbaum 447 31 Huilgol SC, Ramnarain N, Carrington P, Leigh IM, Black 50 Gertz MA, Garton JP, Kyle RA. Primary amyloidosis (AL) in MM. Cytokeratins in primary cutaneous amyloidosis. families. Am J Hematol 1986; 22: 193–198. Australas J Dermatol 1998; 39: 81–85. 51 Landgren O, Gridley G, Turesson I, Caporaso NE, Goldin LR, 32 Centola M, Aksentijevich I, Kastner DL. The hereditary Baris D et al. Risk of monoclonal gammopathy of undeter- periodic fever syndromes: molecular analysis of a new family mined significance (MGUS) and subsequent multiple mye- of inflammatory diseases [Review]. Hum Mol Genet 1998; 7: loma among African-American and White veterans in the 1581–1588. US. Blood 2006; 107: 904–906. 33 Saraiva MJ, Costa PP, Goodman DS. Studies on plasma 52 Buck FS, Koss MN, Sherrod AE, Wu A, Takahashi M. Ethnic transthyretin (prealbumin) in familial amyloidotic polyneuro- distribution of amyloidosis: an autopsy study. Mod Pathol pathy, Portuguese type. J Lab Clin Med 1983; 102: 590–603. 1989; 2: 372–377. 34 Benson MD, Liepnieks J, Uemichi T, Wheeler G, Correa R. 53 Schaffer J, Floege J, Koch KM. Clinical aspects of dialysis- Hereditary renal amyloidosis associated with a mutant related amyloidosis. Contr Nephrol 1995; 112: 90–96. fibrinogen a-chain. Nat Genet 1993; 3: 252–255. 54 Koda Y, Nishi S, Miyazaki S, Haginoshita S, Sakurabayashi T, 35 Levy E, Haltia M, Fernandez-Madrid I, Koivunen O, Ghiso J, Suzuki M et al. Switch from conventional to high-flux Prelli F et al. Mutation in gelsolin gene in Finnish hereditary membrane reduces the risk of carpal tunnel syndrome amyloidosis. J Exp Med 1990; 172: 1865–1867. and mortality of hemodialysis patients. Int 1997; 52: 36 Chartier-Harlin MC, Crawford F, Houlden H, Warren A, 1096–1101. Hughes D, Fidani L et al. Early-onset Alzheimer’s disease 55 Schwalbe S, Holzhauer M, Schaeffer J, Galanski M, Koch KM, caused by mutations at codon 717 of the b-amyloid precursor Floege J. Beta 2-microglobulin associated amyloidosis: a protein gene. Nature 1991; 353: 844–846. vanishing complication of long-term hemodialysis? Kidney 37 Nichols WC, Dwulet FE, Liepnieks J, Benson MD. Variant Int 1997; 52: 1077–1083. apolipoprotein AI as a major constituent of human heredi- 56 Gertz MA, Kyle RA. Secondary systemic amyloidosis: tary amyloid. Biochemical and Biophysical Research Commu- response and survival in 64 patients. Medicine 1991; 70: nications 1988; 156: 762–768. 246–256. 38 Pepys MB, Hawkins PN, Booth DR, Vigushin DM, Tennent 57 Chait A, Han CY, Oram JF, Heinecke JW. Thematic review GA, Soutar AK et al. Human lysozyme gene mutations cause series: the immune system and atherogenesis. - hereditary systemic amyloidosis. Nature 1993; 362: 553–557. associated inflammatory proteins: markers or mediators of 39 Cohen DH, Feiner H, Jensson O, Frangione B. Amyloid fibril cardiovascular disease? JLipidRes2005; 46: 389–403. in hereditary cerebral hemorrhage with amyloidosis 58 van der Westhuyzen DR, Cai L, de Beer MC, de Beer FC. (HCHWA) is related to the gastroentero-pancreatic neuroen- Serum amyloid A promotes efflux mediated by docrine protein, gamma trace. J Exp Med 1983; 158: 623–628. scavenger receptor B-I. J Biol Chem 2005; 280: 35890–35895. 40 Rogaev EI, Sherrington R, Rogaeva EA, Levesque G, Ikeda M, 59 Manley PN, Ancsin JB, Kisilevsky R. Rapid recycling of Liang Y et al. Familial Alzheimer’s disease in kindreds with cholesterol: the joint biologic role of C-reactive protein and missense mutations in a gene on chromosome 1 related to the serum amyloid A. Med Hypotheses 2006; 66: 784–792. Alzheimer’s disease type 3 gene. Nature 1995; 376: 775–778. 60 Poitou C, Viguerie N, Cancello R, De Matteis R, Cinti S, Stich 41 Kovacs DM, Tanzi RE. Monogenic determinants of familial V et al. Serum amyloid A: production by human white Alzheimer’s disease: presenilin-1 mutations. Cell Mol Life Sci adipocyte and regulation by obesity and nutrition. Diabeto- 1998; 54: 902–909. 48 42 Citron M, Westaway D, Xia W, Carlson G, Diehl T, Levesque logia 2005; : 519–528. 61 Cohen AS. Amyloidosis associated with rheumatoid arthritis. G et al. Mutant of Alzheimer’s disease increase 52 production of 42-residue -protein in both Med Clin N A 1968; : 643–653. Rheumatology transfected cells and transgenic mice. Nat Med 1997; 3: 67–72. 62 Buxbaum JN. The Amyloidoses. 2003: 43 Buxbaum JN. Mechanisms of disease: monoclonal immuno- 2015–2026. globulin deposition. Hemat/Oncol Clin NA 1993; 6: 323–346. 63 Gillmore JD, Lovat LB, Persey MR, Pepys MB, Hawkins PN. 44 Kyle RA, Lust JA. Monoclonal gammopathies of undeter- Amyloid load and clinical outcome in AA amyloidosis in mined significance. In: Wiernik PH, Canellos GP, Kyle RA, relation to circulating concentration of serum amyloid A Schiffer CA (eds) Neoplastic Diseases of the Blood. New York: protein. Lancet 2001; 358: 24–29. Churchill Livingstone, 1991, pp 571–594. 64 Sipe J. Revised nomenclature for serum amyloid A (SAA). 45 Avet-Loiseau H, Facon T, Grosbois B, Magrangeas F, Rapp MJ, Nomenclature Committee of the International Society of Harousseau JL et al. Oncogenesis of multiple myeloma: 14q32 Amyloidosis. Part 2 [editorial]. Amyloid 1999; 6: 67–70. and 13q chromosomal abnormalities are not randomly 65 Kluve-Beckerman B, Naylor SL, Marshall A, Gardner JC, distributed, but correlate with natural history, immunological Shows TB, Benson MD. Localization of human SAA gene(s) features, and clinical presentation. Blood 2002; 99: 2185–2191. to and detection of DNA polymorphisms. 46 Stevens FJ. Four structural risk factors identify most fibril- Biochem Biophys Res Commun 1986; 137: 1196–1204. forming kappa light chains. Amyloid 2000; 7: 200–211. 66 Qiu H, Shimada K, Cheng Z. Chromosomal localization of 47 Abraham RS, Geyer SM, Price-Troska TL, Allmer C, Kyle RA, the mouse prealbumin gene (Ttr)byin situ hybridization. Gertz MA et al. Immunoglobulin light chain variable (V) Cytogenet Cell Genet 1992; 61: 186–188. region genes influence clinical presentation and outcome in 67 Moriguchi M, Terai C, Koseki Y, Uesato M, Nakajima A, light chain associated amyloidosis (AL). Blood 2003; 101: Inada S et al. Influence of genotypes at SAA1 and SAA2 loci 3801–3807. on the development and the length of latent period of 48 Grosbois B, Jego P, Attal M, Payen C, Rapp MJ, Fuzibet JG secondary AA-amyloidosis in patients with rheumatoid et al. Familial multiple myeloma: report of fifteen families. Br arthritis. Hum Genet 1999; 105: 360–366. J Haematol 1999; 105: 768–770. 68 Booth DR, Booth SE, Gillmore JD, Hawkins PN, Pepys MB. 49 Miliani A, Bergesio F, Salvadori M, Ciciani AM, Merlini GP, SAA1 alleles as risk factors in reactive systemic AA Di Guglielmo R et al. Systemic amyloidosis and monoclonal amyloidosis. Amyloid 1998; 5: 262–265. gammopathy in three Italian siblings: a familian case of AL- 69 Yamada T, Okuda Y, Takasugi K, Wang L, Marks D, Benson amyloidosis?. In: Natvig JB, Fo¨rre O¨ , Husby G, Husebekk A, MD et al. An allele of associated with Skogen B, Sletten K, Westermark P (eds) Amyloid and amyloidosis in both Japanese and Caucasians. Amyloid 2003; Amyloidosis: VIth International Symposium on Amyloidosis, 5–8 10: 7–11. August 1990, Oslo, Norway. Kluwer Academic Publishers: 70 Okuda Y, Yamada T, Takasugi K, Takeda M, Nanba S, Onishi Dordrecht, 1990, pp 227–230. M et al. Serum amyloid A (SAA) 1, SAA 2 and apolipoprotein

Genes and Immunity The genetics of the amyloidoses J Buxbaum 448 E isotype frequencies in rheumatoid arthritis patients with familial Mediterranean fever homozygous for the M694V- AA amyloidosis]. Ryumachi 1999; 39: 3–10. MEFV mutation. J Rheumatol 2003; 30: 308–312. 71 Nakamura T, Higashi S, Tomoda K, Tsukano M, Baba S, 88 Cazeneuve C, Ajrapetyan H, Papin S, Roudot-Thoraval F, Shono M. Significance of SAA1.3 allele genotype in Japanese Genevieve D, Mndjoyan E et al. Identification of MEFV- patients with amyloidosis secondary to rheumatoid arthritis. independent modifying genetic factors for familial Mediter- Rheumatology (Oxford) 2005; 84: 1322–1328. ranean fever. Am J Hum Genet 2000; 67: 1136–1143. 72 Ishii W, Matsuda M, Nakamura A, Nakamura N, Suzuki A, 89 van der Hilst JC, Drenth JP, Bodar EJ, Bijzet J, van der Meer Ikeda S. Abdominal fat aspiration biopsy and genotyping of JW, Simon A. Serum amyloid A serum concentrations and serum amyloid A contribute to early diagnosis of reactive genotype do not explain low incidence of amyloidosis in AA amyloidosis secondary to rheumatoid arthritis. Intern Hyper-IgD syndrome. Amyloid 2005; 12: 115–119. Med 2003; 42: 800–805. 90 Richter GW. The resorption of amyloid under experimental 73 Moriguchi M, Kaneko H, Terai C, Koseki Y, Kajiyama H, conditions. Am J Pathol 1954; 30: 239–261. Inada S et al. Relative transcriptional activities of SAA1 91 Durie BG, Persky B, Soehnlen BJ, Grogan TM, Salmon SE. promoters polymorphic at position À13(T/C): potential Amyloid production in human myeloma stem-cell culture, association between increased transcription and amyloidosis. with morphologic evidence of amyloid secretion by asso- Amyloid 2005; 12: 26–32. ciated macrophages. N Engl J Med 1982; 307: 1689–1692. 74 Yamada T, Okuda Y, Itoh Y. The frequency of serum amyloid 92 Keeling J, Teng J, Herrera GA. AL-amyloidosis and light- A2 alleles in the Japanese population. Amyloid 1998; 5: 208– chain deposition disease light chains induce divergent 211. phenotypic transformations of human mesangial cells. Lab 75 Faulkes DJ, Woo P. Do alleles at the serum amyloid A locus Invest 2006; 45: 43–49. influence susceptibility to reactive amyloidosis in systemic 93 Hrncic R, Wall J, Wolfenbarger DA, Murphy CL, Schell M, onset juvenile chronic arthritis? Amyloid:Int J Exp Clin Invest Weiss DT et al. Antibody-mediated resolution of light 1979; 4: 75–79. chain-associated amyloid deposits. Am J Pathol 2000; 157: 76 Stojanov S, Kastner DL. Familial autoinflammatory diseases: 1239–1246. genetics, pathogenesis and treatment. Curr Opin Rheumatol 94 Solomon A, Weiss DT, Wall JS. in systemic 2005; 17: 586–599. primary (AL) amyloidosis using amyloid-reactive monoclo- 77 Aganna E, Martinon F, Hawkins PN, Ross JB, Swan DC, nal antibodies. Cancer Biother Radiopharm 2003; 18: 853–860. Booth DR et al. Association of mutations in the NALP3/ 95 Bales KR, Du Y, Dodel RC, Yan GM, Hamilton-Byrd E, Paul CIAS1/PYPAF1 gene with a broad phenotype including SM. The NF-kappaB/Rel family of proteins mediates Abeta- recurrent fever, cold sensitivity, sensorineural deafness, and induced neurotoxicity and glial activation. Brain Res Mol AA amyloidosis. Arthritis Rheum 2002; 46: 2445–2452. Brain Res 1998; 57: 63–72. 78 Hull KM, Drewe E, Aksentijevich I, Singh HK, Wong K, 96 Yan SD, Chen X, Fu J, Chen M, Zhu H, Roher A et al. RAGE McDermott EM et al. The TNF receptor-associated periodic and amyloid-b peptide neurotoxicity in Alzheimer’s disease. syndrome (TRAPS): emerging concepts of an autoinflamma- Nature 1996; 382: 685–692. tory disorder. Medicine (Baltimore) 2002; 81: 349–368. 97 El Khoury JE, Hickman SE, Thomas CA, Cao L, Silverstein 79 Feldmann J, Prieur AM, Quartier P, Berquin P, Cortis E, SC, Loike JD. Scavenger receptor-mediated adhesion of Teillac-Hamel D et al. Chronic infantile neurological cuta- microglia to b-amyloid fibrils. Nature 1996; 382: 716–719. neous and articular syndrome is caused by mutations in 98 Kunjathoor VV, Tseng AA, Medeiros LA, Khan T, Moore KJ. CIAS1, a gene highly expressed in polymorphonuclear cells Beta-amyloid promotes accumulation of lipid peroxides by and chondrocytes. Am J Hum Genet 2002; 71: 198–203. inhibiting CD36-mediated clearance of oxidized lipoproteins. 80 Dode C, Le Du N, Cuisset L, Letourneur F, Berthelot JM, J 2004; 1: 23. Vaudour G et al. New mutations of CIAS1 that are 99 Schenk D, Barbour R, Dunn W, Gordon G, Grajeda H, Guido responsible for muckle–wells syndrome and familial cold T et al. Immunization with amyloid-beta attenuates Alzhei- urticaria: a novel mutation underlies both syndromes. Am J mer-disease-like pathology in the PDAPP mouse. Nature Hum Genet 2002; 70: 1498–1506. 1999; 400: 173–177. 81 Dode C, Andre M, Bienvenu T, Hausfater P, Pecheux C, 100 Broytman O, Malter JS. Anti-Abeta: the good, the bad, and Bienvenu J et al. The enlarging clinical, genetic, and the unforeseen. J Neurosci Res 2004; 75: 301–306. population spectrum of tumor necrosis factor receptor- 101 Monsonego A, Imitola J, Petrovic S, Zota V, Nemirovsky A, associated periodic syndrome. Arthritis Rheum 2002; 46: Baron R et al. Abeta-induced is IFN- 2181–2188. gamma-dependent and is associated with T cell-dependent 82 Chae JJ, Komarow HD, Cheng J, Wood G, Raben N, Liu PP clearance of Abeta in a mouse model of Alzheimer’s disease. et al. Targeted disruption of pyrin, the FMF protein, causes Proc Natl Acad Sci USA 2006; 103: 5048–5053. heightened sensitivity to endotoxin and a defect in macro- 102 Nicoll JA, Wilkinson D, Holmes C, Steart P, Markham H, phage . Mol Cell 2003; 11: 591–604. Weller RO. Neuropathology of human Alzheimer disease 83 McDermott MF. Genetic clues to understanding periodic after immunization with amyloid-beta peptide: a case report. fevers, and possible therapies. Trends Mol Med 2002; 8: Nat Med 2003; 9: 448–452. 550–554. 103 Lee M, Bard F, Johnson-Wood K, Lee C, Hu K, Griffith SG 84 Jacobson DR, Buxbaum JN. Genetic aspects of amyloidosis. et al. Abeta42 immunization in Alzheimer’s disease generates Adv Hum Genet 1991; 20: 69–123. Abeta N-terminal antibodies. Ann Neurol 2005; 58: 430–435. 85 Gershoni-Baruch R, Brik R, Zacks N, Shinawi M, Lidar M, 104 Bard F, Cannon C, Barbour R, Burke RL, Games D, Grajeda H Livneh A. The contribution of genotypes at the MEFV and et al. Peripherally administered antibodies against amyloid SAA1 loci to amyloidosis and disease severity in patients beta-peptide enter the central nervous system and reduce with familial Mediterranean fever. Arthritis Rheum 2003; 48: pathology in a mouse model of Alzheimer disease. Nat Med 1149–1155. 2000; 6: 916–919. 86 Akar N, Hasipek M, Akar E, Ekim M, Yalcinkaya F, Cakar N. 105 Terazaki H, Ando Y, Fernandes R, Yamamura KI, Maeda S, Serum amyloid A1 and tumor necrosis factor-alpha alleles in Saraiva MJ. Immunization in familial amyloidotic polyneuro- Turkish familial Mediterranean fever patients with and pathy: counteracting deposition by immunization with a without amyloidosis. Amyloid 2003; 10: 12–16. Y78F TTR mutant. Lab Invest 2006; 86: 23–31. 87 Gershoni-Baruch R, Brik R, Lidar M, Shinawi M, Livneh A. 106 Sekijima Y, Wiseman RL, Matteson J, Hammarstrom P, Miller Male sex coupled with articular manifestations cause a 4-fold SR, Sawkar AR et al. The biological and chemical basis for increase in susceptibility to amyloidosis in patients with tissue-selective amyloid disease. Cell 2005; 121: 73–85.

Genes and Immunity The genetics of the amyloidoses J Buxbaum 449 107 Sousa A, Coelho T, Lobato L, Sequeiros J. Genetic epidemio- complexity in a single-gene disease. Hum Mol Genet 2005; 14: logy of familial amyloidotic polyneuropathy (FAP)-type I 543–553. (FAP-MET30). In: Kyle R (ed). Amyloid and Amyloidosis. 110 Casadei VM, Ferri C, Veglia F, Gavazzi A, Salani G, Cattaneo Parthenon Publishers: New York and London, 1995, M et al. APOE-491 promoter polymorphism is a risk factor for pp 209–211. late-onset Alzheimer’s disease. Neurology 1999; 53: 1888–1889. 108 Sousa A, Andersson R, Drugge U, Holmgren G, Sandgren O. 111 Corder EH, Saunders AM, Strittmatter WJ, Schmechel DE, Familial amyloidotic polyneuropathy in Sweden: geographi- Gaskell PC, Small GW et al. Gene dose of cal distribution, age of onset, and prevalence. Hum Hered type 4 allele and the risk of Alzheimer’s disease in late onset 1993; 43: 288–294. families. Science 1993; 261: 921–923. 109 Soares ML, Coehlo T, Sousa A, Batalov S, Conceicao I, 112 Hoshii Y, Kawano H, Cui D, Takeda T, Gondo T, Takahashi M Sales-Luis ML et al. Susceptibility and modifier genes in et al. Amyloid A protein amyloidosis induced in apolipopro- Portuguese transthyretin V30M amyloid polyneuropathy: tein-E-deficient mice. Am J Pathol 1997; 151: 911–917.

Genes and Immunity