Prion disease

Abstract

Genetic prion diseases (gPrDs) are caused by autosomal-dominant mutations in the prion protein gene (PRNP). Although the first PRNP mutations identified, and most since, are PRNP missense, octapeptide repeat insertions, deletion and nonsense mutations have now also been shown to cause gPrD. Based on clinicopathologic features of familial disease, gPrDs historically have been classified into three forms: familial Jakob–Creutzfeldt disease, Gerstmann–Sträussler–Scheinker disease, and fatal familial insomnia. This classification, however, occurred prior to the identification of PRNP, and although these forms are still recognized, classification now is somewhat more complex. Clinical manifestations, and even pathology, are known to be more heterogeneous and varied than the historic three phenotypic classifications. Most gPrDs either present rapidly with progression of dementia, ataxia, myoclonus, and other motor features leading to death in few months or present more slowly, declining over a few years with mild cognitive impairment, ataxia, or parkinsonism and later dementia; a few very rare mutations, however, present over years to decades with neuropsychiatric disorders and systemic symptoms (gastrointestinal disorders and neuropathy). In this chapter, we review the broad phenotypic spectrum of PRNP mutations causing gPrDs.

INTRODUCTION

Prion diseases (PrDs) are neurodegenerative disorders caused by misfolded forms of the prion protein (PrP) called “prions,” derived from the term proteinaceous infectious particles (Prusiner, 1982). PrDs occur not only in humans, but also in animals, such as sheep and goats (scrapie) and cattle (bovine spongiform encephalopathy), among others (Prusiner, 1998). Human PrDs occur in three ways: sporadic (spontaneous), genetic, and acquired. Sporadic PrD is the most common form (around 85–90%), and is represented by sporadic Jakob–Creutzfeldt disease (sJCD), which also includes two other very rare conditions: variably protease-sensitive prionopathy and sporadic fatal insomnia (Masters et al., 1979; Ladogana et al., 2005; Zou et al., 2010; Puoti et al., 2012). Acquired PrDs (infectious forms) are the rarest form, accounting for less than 1% of human PrDs, and include kuru, iatrogenic JCD and variant JCD (Masters et al., 1979; Ladogana et al., 2005; Geschwind, 2015). Genetic PrDs (gPrDs) account for the remaining 10–15% of human PrDs and are caused by autosomal-dominant mutations in PRNP, the gene that encodes the PrP. Historically, gPrDs have been divided into three clinicopathologic phenotypes: familial JCD (fJCD), Gerstmann–Sträussler–Scheinker (GSS) disease, and familial fatal insomnia (FFI). This classification developed, however, prior to the identification of PRNP (Masters et al., 1979), and some gPrDs do not fit well into this categorization (Ladogana et al., 2005).

Regarding terminology in this chapter, we use the term Jakob–Creutzfeldt disease (JCD) rather than Creutzfeldt–Jakob disease, as Hans Creutzfeldt’s case did not actually have what today is considered PrD, whereas at least two of five of Alfons Jakob’s cases did (Masters, 1989). For many decades the terms JCD and Jakob disease were used, but C.J. Gibbs, a prominent researcher in the field, began using the term Creutzfeldt–Jakob disease, so that the acronym, CJD, would be closer to his own initials (Gibbs, 1992). Therefore, JCD or even Jakob disease are the more appropriate terms (Katscher, 1998). Furthermore, as not all patients with PRNP mutations have affected family members (such as due to reduced penetrance or de novo mutations), the terms familial, hereditary, and inherited are often not appropriate. Thus, in this chapter, generally we will use the terms genetic JCD (gJCD) and genetic prion disease (gPrD).

For decades, much of the scientific community mistakenly considered PrDs to be caused by slow viruses (Prusiner, 1998). Although familial aggregation of some PrDs had been reported since the early 1900s, the famili- ality was often attributed to “the virus” being passed down in a family. In the 1980s and early 1990s, Dr. Stanley Prusiner and others showed that PrDs were not caused by viruses or bacteria, but rather by a misfolded form of normal cellular protein, PrP. It was only in 1989 that the PrP gene, PRNP, was identified and familial PrDs were proven to be genetic, due to PRNP mutations (Goldgaber et al., 1989; Hsiao et al., 1989; Owen et al., 1989).

The first autopsied case of gPrD may have been a British patient who presented at age 38 with severe depression and abnormal behavior, followed by memory impairment, and who died in 1901 at the age of 44. Brain pathology at that time was only reported as showing atrophy; pathologic features of PrD were defined later. Almost 90 years later, after the discovery of the PrP gene, a PRNP 6-octapeptide repeat insertion (6-OPRI) mutation was found in this British family (Mead, 2006). fJCD was first reported formally by Meggendorfer in 1930 in a German kindred (the Backer family) that later was shown to carry a missense D178N PRNP mutation (Kretzschmar et al., 1995). In 1936, J. Gerstmann, E. Sträussler, and I. Scheinker reported an Austrian kindred (the “H” family) “with a strange hereditary disease of the central nervous system”–the first known GSS family. The P102L PRNP mutation was eventually identified in this family (Hainfellner et al., 1995). Lugaresi and colleagues used the term familial fatal insomnia (FFI) first in 1986 when describing an Italian family with progressive insomnia, dysautonomia, and motor signs. The neuropathologic findings suggested that FFI was a form of familial PrD, but the confirmation only came in 1992, when a PRNP D178N mutation was identified in the kindred (Medori et al., 1992).

PRION PROTEIN STRUCTURE, FUNCTION, AND GENE (PRNP)

PrP structure function

PRNP is located on chromosome 20p13 and is composed of two exons, but the gene’s entire open reading frame is located in exon 2. The primary sequence of the major PrP has 253 amino acids before posttranslational modification and contains an unstable region in the N-terminal domain, called the octapeptide repeat region, composed of a nonapeptide (R1) followed by four repeats of octapeptides (R2, R2, R3, R4) (PRNP, shown in Fig. 29.1). Posttranslational modifications of PrP include removal of 22 residues from the N-terminal and 23 residues from the C-terminal, along with the attachment of a glycosyl-phosphatidylinositol (GPI) anchor in the C-terminal. A globular domain extends from residues 125 to 228, and contains three α-helices (residues 144–154, 173–194, and 200–228) and an antiparallel β-sheet (residues 128–131 and 161–164) (Zahn et al., 2000). PrP can be glycosylated at asparagine residues 181 and 197 resulting in un-, mono-, or di-glycosylated forms (Capellari et al., 2011; Yusa et al., 2012). The tertiary structure of PrP is influenced by hydrophobic and aromatic interactions between residues of the globular domain, as well as by the existence of salt and also disulfide bridges (van der Kamp and Daggett, 2009; Capellari et al., 2011; Biljan et al., 2013).

Fig. 29.1.

Schematic of PRNP disease-associated variants. Mutations are color coded based on clinicopathologic classifcation as genetic Jakob–Creutzfeldt disease (gJCD), Gerstmann–Sträaussler–Scheinker (GSS), fatal familial insomnia (FFI), or nonsense mutations. PRNP mutations present in the University of California, San Francisco cohort are in bold. Most mutations are shown below the gene schematic; nonsense mutations and polymorphisms associated with prion disease risk are above the gene schematic. Low- or intermediate-penetrance variants are based on Minikel et al. (2016) (not all low/intermediate-penetrance variants are shown). For the F198V mutation, the clinical presentation was not classifiable as gJCD, GSS, or FFI (see Table 29.3), and neuropathology was not reported (Zheng et al., 2008). Variants that are probably benign (largely based on Minikel et al., 2016) are not included (e.g., G54S, P39L, E196A, R208C) (Beck et al., 2010; Minikel et al., 2016). OPRI, octapeptide repeat insertion; OPRD, octapeptide repeat deletion. (Reproduced from Takada et al. (2017), with permission from John Wiley.)

The normal, cellular form of PrP is often referred to as PrP^C, in which “C” stands for cellular. The misfolded, disease-causing, form of the PrP, called the prion, is commonly referred to as PrP^Sc, in which “Sc” stands for the PrD of sheep and goats, or sometimes as PrP^Res, as the prion is partially “resistant” to proteases that completely degrade PrP^C (Prusiner, 1998). PrP^C is typically attached to the cell plasma membrane by a GPI anchor, but PrP^C is also internalized and cycles between the plasma membrane and endosomes (Harris, 2003).

The functional role of PrP^C is still not completely clear. PRNP is highly conserved across mammals, indicating PrP was important during evolution (Colby and Prusiner, 2011). The octapeptide repeat domain in the N-terminal region contains copper-binding sites, which suggests PrP^C might have a role in the copper homeostasis (Harris et al., 2003). Many other potential functions of PrP^C have been suggested, including roles in brain development, adult neural plasticity, neurotransmission, neuroprotection, receptors for proteins such as beta-amyloid and, most recently, myelin maintenance (Flechsig and Weissmann, 2004; Watts and Westaway, 2007; Aguzzi et al., 2008; Biasini et al., 2012; Bribian et al., 2012; Didonna, 2013; Kuffer et al., 2016; Slapsak et al., 2016).

Both PrP^C and PrP^Sc share the same amino acid sequence, but whereas PrP^C structure is composed mainly of alpha-helices, PrP^Sc has a high beta-sheet content (Pan et al., 1993; Wille et al., 2009; Colby and Prusiner, 2011). This conformational change affects PrP biochemical properties, so that PrP^Sc becomes insoluble in detergents and relatively resistant to degradation by proteases (Mastrianni, 2010; Colby and Prusiner, 2011). PrP^Sc self-propagates by acting as a template to cause the misfolding of PrP^C into PrP^Sc (Prusiner, 1998). Although it is not entirely clear where in the cell this templated misfolding occurs, there is evidence suggesting it occurs inside cholesterol-rich nonacidic intracellular compartments called caveolae-like domains (Colby and Prusiner, 2011). In gPrDs, it is generally thought that PRNP mutations make PrP^C proteins more susceptible to misfolding into PrP^Sc.

Researchers disagree on whether it is the conversion of PrP^C to PrP^Sc or the accumulation of the latter that leads to neuronal injury and death (Prusiner, 1998; Verity and Mallucci, 2011; Moreno et al., 2013). PrP^Sc propagates as oligomers, which might polymerize to form amyloid fibrils (Prusiner, 2013). As noted below, PrP^Sc amyloid plaques are found in brains of some forms of PrD, particularly GSS, but also some subtypes of sJCD and iatrogenic JCD.

There are two major types of PrP^Sc found in brains of individuals with PrD, termed types 1 and 2 (as proposed by Gambetti and colleagues, 2003). After brain homogenates are treated with proteinase K and PNGase, PrP^Sc type 1 shows up as an unglycosylated band at21 kDa and type 2 as an unglycosylated band at 19 kDa on Western blot analysis (Parchi et al., 1996). Most patients only have one PrP^Sc type, but co-occurrence of types 1 and 2 is found in about a third of sJCD cases (Parchi et al., 2009). This PrP^Sc type combined with PRNP codon 129 polymorphism has been as a molecular classification scheme for sporadic JCD (see Puoti et al. (2012) for a review and Hill et al. (2003)).

PRNP variants and pathogenic mechanisms

PRNP sequence variants that cause gPrD include missense mutations, OPRIs/octapeptide repeat deletions (OPRDs), and nonsense mutations. Approximately 50 mutations associated with gPrD have been identified (Takada et al., 2017). The locations of most known PRNP mutations are shown schematically in Figure 29.1.

PRNP missense mutations

The majority of PRNP mutations are missense, with more than 35 pathogenic missense mutations reported (Takada et al., 2017). There is no single pathogenic mechanism underlying missense mutations, which is probably a major reason for the striking heterogeneity seen across gPrD. Missense mutations can cause gJCD, GSS, or FFI.

PRNP octapeptide repeat insertions and deletions

Insertions and deletions of repeats in the octapeptide repeat region (R1-R2-R2-R3-R4) between codons 51 and 91 of PRNP (Fig. 29.1) can also cause gPrD (Owen et al., 1989). Most insertions occur between the second R2 and R3 (Capellari et al., 2002; Jansen et al., 2011). OPRIs and OPRDs are usually classified based on the number of additional repeats found, such as 6-OPRI having six extra repeats. An insertion or deletion of one octapeptide is not considered pathogenic, as 1-OPRI and 1-OPRD have also been found in many healthy individuals (Capellari et al., 2002; Yu et al., 2004; Beck et al., 2010). 2-OPRD, on the other hand, was reported in a few cases and is considered pathogenic (Capellari et al., 2002). OPRIs have been reported with 2–12 extra OPRIs and show a wide spectrum of clinical and neuropathologic phenotypes (discussed below). Complicating matters, however, is that octapeptide mutations of the same number are not always equivalent between families, as the exact position of the insertion/deletion is not always the same and there might be minor nucleotide sequence variations between families (Croes et al., 2004). For example, unrelated families with 7-OPRI could have different combinations of seven repeats (e.g., R2–R2–R2–R2–R3g–R2–R2 in one family vs. R3–R2–R3–R2–R2–R2–R2 in another), which might in part explain clinicopathologic phenotypic differences between families (Jansen et al., 2011).

There are several potential mechanisms for how PRNP octapeptide insertions (or deletions) might cause gPrD. PrP^C with two or more extra octapeptide repeats has been shown to form protease-resistant PrP more rapidly, which appears to be due to more rapid binding between PrP molecules (PrP^C–PrP^Sc) rather than PrP conformational change from extra octapeptide repeats (Priola and Chesebro, 1998; Moore et al., 2006). Furthermore, the octapeptide region of PRNP binds copper ions and OPRIs appear to alter copper binding, which might cause either loss of PrP^C copper-mediated function or loss of copper-dependent protection against conversion to PrP^Sc (Leliveld et al., 2006; Hodak et al., 2009; Stevens et al., 2009).

PRNP nonsense mutations

PRNP nonsense (or frameshift mutation leading to premature stop codon) mutations are very rare and cause gPrD with atypical clinical and neuropathologic features, as discussed below. The pathogenic mechanisms underlying these mutations are still unclear, but the lack of the GPI anchor in the truncated protein appears to play an important role (Mead et al., 2013). Experiments with transgenic mice expressing anchorless PrP showed that not only can they develop a transmissible PrP amyloidosis, but also that following infection by PrP^Sc, PrP deposition can be found in extraneural tissues such as heart, kidney, pancreas and gut (Stohr et al., 2011).

PRNP polymorphisms in PRD

As shown in Figure 29.1, there are several polymorphisms in PRNP. Codon 129 (refSNP rs1799990) is the most important one as it has a strong influence on both disease susceptibility and the phenotypical presentation of PrDs. Codon 129 encodes either methionine or valine, and in the general population of European ancestry, about 55% of persons are homozygous for methionine (MM), 36% are heterozygous (MV), and 9% are homozygous for valine (VV) (1000 Genomes Project Consortium et al., 2012). Homozygosity at codon 129 is a well-established risk factor for sporadic and acquired PrDs (Palmer et al., 1991). Amongst sJCD cohorts from Europe, the United States, Canada, and Australia, for example, 67% are MM, 16% MV, and 17% VVat codon 129 (Parchi et al., 1999; Collins et al., 2006). Amongst Asians, however, valine is very uncommon at codon 129 and thus most are MM with very few MV and none are VV (Nozaki et al., 2010; Moe Lee et al., 2012). The rate of progression of sJCD appears to be strongly affected by codon 129 (Mead et al., 2016).

As the conversion of PrP^C into PrP^Sc is thought to be facilitated by the presence of proteins with identical amino acid sequences, homozygotes would have more substrate for PrP^Sc formation than heterozygotes and hence would theoretically be more susceptible to PrD (Palmer et al., 1991). The residue 129 does not alter PrP^C conformation or stability, but seems to mediate protein– protein contact during PrP^Sc propagation (Hosszu et al., 2004; Antonyuk et al., 2009). Moreover, the codon 129 polymorphism has been shown to affect the selection and propagation of prion strains, as some types of PrP^Sc are only observed in certain genotypes (Wadsworth et al., 2004). There is also evidence that codon 129 polymorphism modulates the formation of amyloid fibrils, which is another potential mechanism underlying the phenotypical changes associated with this polymorphism (Apetri et al., 2005).

In gPrD, the codon 129 polymorphism located on the same allele as the mutation (cis) has a more significant effect on disease presentation, whereas the trans polymorphism has less (or possibly no) effect (Capellari et al., 2011). A powerful example of the cis effect of codon 129 is observed in those carrying a D178N mutation. When D178N gPrD is associated with cis M, patients usually present with an FFI clinicopathologic phenotype (see below), whereas D178N with cis V usually present as an sJCD clinicopathologic phenotype (Goldfarb et al., 1992; Zarranz et al., 2005).

The codon 129 polymorphism is also a strong risk factor for acquired JCD. Almost all symptomatic cases of variant JCD reported to date have been homozygous for methionine, suggesting valine at codon 129 is protective in variant JCD against the development of symptoms (Wadsworth et al., 2004; Bishop et al., 2013). But it was later discovered that valine does not prevent infection, as a few subclinical/preclinical cases of variant JCD have been identified by the detection of PrP^Sc in lymphoreticular tissues of individuals with the MV and VV genotypes (Gill et al., 2013) and a pathology-proven MV case was recently identified (Mok et al., 2017). Homozygosity at codon 129 also is linked to an increased risk of iatrogenic JCD due to human growth hormones and kuru (Brandel et al., 2003; Bishop et al., 2009).

Other PRNP risk variants have been described, though only in selected populations. Amongst Asians, particularly Japanese and Koreans, there is polymorphism at PRNP codon 219, resulting in an amino acid substitution of glutamate (E) for lysine (K). The allelic frequency of 219K in this Asian population is 4–8% (Jeong et al., 2004; Bishop et al., 2009), but it is not found in Caucasian populations. Heterozygosity at codon 219 (EK) is protective against sJCD compared to homozygosity (EE) (Shibuya et al., 1998). This polymorphism apparently has not shown an effect in acquired PrD, but seems to affect the clinicopathologic features of GSS due to the P102L mutation (Tanaka et al., 1997). The presence of lysine at codon 219 seems to prevent PrP^C from being converted into PrP^Sc in sJCD models (Kaneko et al., 1997; Perrier et al., 2002). Interestingly, heterozygosity at this polymorphism appears to be a risk factor for variant JCD (Bishop et al., 2009; Lukic et al., 2010).

Heterozygosity at codon 127 with a glycine being substituted for a valine (127GV) in PRNP has shown to be protective against kuru in the Fore ethnic group of Papua New Guinea (Mead et al., 2009). Furthermore, when this valine polymorphism was put into humanized mice, they were protected against PrD. PrP^C with 127V not only could not be converted to PrP^Sc, but it appeared to act as a dose-dependent inhibitor of prion propagation (Asante et al., 2015). This may suggest a potential mechanism for treating or preventing PrD.

Purported mutations that might just be polymorphisms with strong risk factors

Increasing evidence suggests that a few of the 50-plus PRNP sequence variants, such as M232R, probably are not causative mutations, but rather risk factors or very low-penetrance mutations (Minikel et al., 2016). Some PRNP variants have a very low or even absent family history of PrD. A recent paper using data from the Exome Aggregation Consortium and 23andMe showed that some PRNP variants were much more frequently found in the general population than expected for Mendelian disease-causing, highly penetrant variants. Thus, these variants are very likely low-penetrance mutations or risk factors or just variants irrelevant to PrD risk (Minikel et al., 2016). Although the V210I mutation is the most frequently reported mutation in Italy, it only has penetrance of about 10%. V180I and M232R, mutations commonly found in Japan have much lower estimated penetrance of 1% and 0.1%, respectively (Minikel et al., 2016). Other variants reported only in a few patients and often without a family history of PrD are included in Table 29.1.

Table 29.1.

PRNP missense mutations

PRNP mutation	Codon 129 polymorphism	# of cases in literatureb	Clinical phenotypes	Age at onset (range)f (years)	Disease duration (months or years)	Positive family historyc	CSF marker sensitivity		EEG PSWC	MRI c/w JCDd	Neuropathology	Neuropathology phenotype	References
							14-3-3	Total taug
P84S	MV	1	Cog (1 year) → paranoia and RPD	(60)	(14) months	0% (0/1)	N/A		0% (0/1)	0% (0/1)	Multicentric PrP-Plqs w/o NFT No V	GSS	Jones et al. (2014)
S97N	Cis M	1	AD	(72)	N/A	0% (0/1)	N/A		N/A	0% (0/1)	N/A	N/A	Zheng et al. (2008)
P102L	MM/MV (most cis M)	~221	Early Cb w/ late D Some are RPD LE areflexia common	(27–66)	(7–132) months	84–100%	(15–20% )	20%	0–35%	25–30%	Multicentric PrP-Plqs	GSS	Webb et al. (2008); Higuma et al. (2013); Krasnianski et al. (2016)
P102La	Cis M	13		44 ± 12 (24–57)	44 ± 13 months (28–60)	FHx score 0 (n = 1) 1 (n = 2) 2 (n = 6)	0% (0/3)	0%	0% (0/3)	33.3% (1/3)			Takada et al. (2017))
P105L	MV	13	D w/ spastic paraparesis Cb Atx Pscyh Sxs sometimes	Mean 44 ± 10 (2nd to 7th decades)	111 ± 82 months	37%	0%	0%	0%	14%	PrP-Plqs diff PrP (deep CLs)	GSS	Higuma et al. (2013)
P105La	N/A	1		(9)	N/A	FHx Score 0 (n = 1)	N/A	N/A	0% (0/1)	0% (0/1)	N/A		Takada et al. (2017)
P105T	Cis M	13	Usually RPD; Cb Atx freq	(13–41)	(2-5) years	100% (2/2)	0%	N/A	0% (0/3)	25% (1/4)	V, PrP-S (all CLs) Unicentric PrP-Plqs (deep CLs)	JCD	Rogaeva et al. (2006);Polymenidou et al. (2011)
P105S	MV (cis V)	1	Aphasia, frontal-type behavioral changes, D, late Park	(30)	(10) years	0% (0/1)	N/A	N/A	0% (0/1)	100% (1/1)	Multicentric PrP-Plqs (HP), punctate aggregates (Cb), V (Pu)	Atypic GSS	Tunnell et al. (2008)
G114V	(MM/MV)	1	RPD Onset: Psych Sxs, D, Park, Pyram signs, myoclonus GTCs in some Absent or mild Cb signs	(18–75)	(1–4) years	75% (3 in 4 probands) e	0%		0% (0/8)	42.9% (3/7) predom BG	Mod V, G, NL. PrP-S, type 1 PrPSc (predom monoglycos)	JCD	Rodriguez et al. (2005); Ye et al. (2008); Beck et al., (2010);Liu et al. (2010)
A117V	Cis V	33	Variable Progressive D w/o Atx LMN SYN w/ D and Atx	(20–64)	(1–11 years)	100% (4/4)	N/A	N/A	N/A	0% (0/2)	Ab PrP-Plqs, F V, NL, G	GSS	Hsiao et al. (1991); Mastrianni et al.(1995); Kong et al. (2004)
A117V*	Cis V	6		34 ± 14 (14–49)	46 ± 21 months (27–78)	FHx score 0 (n = 1) 2 (n = 2)	0%	0%	0% (0/3)	0% (0/5)			Takada et al. (2017)
G131V	MM/MV (cis M)	3	D w/ behavioral changes and late Atx Park	(36–42)	(9–16) years	50% (1/2)	N/A	N/A	0% (0/1)	0% (0/1)	PrP-Plqs, NFT (AH, ERC), No V	GSS	Panegyres et al. (2001); Jansen et al. (2012)
S132I	MM	2	RPD	(62)	(18) months	100% (1/1)	N/A	N/A	N/A	N/A	Diff unicentric and multicentric PrP-PLs (neocortex, BG, Cb) Min V	GSS	Hilton et al. (2009)
A133V	MM	2	PSP-like, RPD	(62)	(4) months	0% (0/1)	0%		0% (0/1)	0% (0/1)	Diff V, G, NL multic PrP-Plqs in the (Mol), PrP-S (Th)	Atypic GSS	Rowe et al. (2007)
R148H	(MV/MM)	3	JCD	(62–82)	(6–18) months	0% (0 in 2)e	50% (1/2)	100% (1/1)	50% (1/2)	100% (1/1) predom BG	129MM - similar to sJCDMM1. V, G, NL (deeper CLs). PrP-S PrPSc type 1 129MV - similar to sJCDMV2 V predom in CLs V and VI, Kuru plaques in Cb and WM, PrP-S PrPSc type 2 (predom monoglycos)	JCD	Krebs et al. (2005); Pastore et al.(2005)
R148H*	MM	1		(53)	(2) months	FHx score 0 (n = 1)	N/A	N/A	N/A	100% (1/1)			Takada et al. (2017)
D167G	MM	1	JCD	N/A	N/A	N/A	N/A	N/A	N/A	N/A	sJCD PrP type 1	JCD	Bishop et al. (2009)
D167N	MM	1	RPD w/ Park and Pyram signs	(33)	(2) years	0% (0/1)e	N/A	N/A	0% (0/1)	0% (0/1)	N/A	N/A	Beck et al. (2010)
V176G	VV	1	RPD w/ behavioral changes, Cb Atx, Pyram signs and myoclonus	(61)	(7) months	0% (0/1)	100% (1/1)	100% (1/1)	0% (0/1)	0% (0/1)	Multic PrP-Plqs w/ prominent tau	GSS	Simpson et al. (2013)
D178N-129V	MV/VV (cis V)	209a	Progressive Cog decline, Cb Sxs, myoclonus, EP Sxs	Mean 46 (26–56)	Mean 23 months (7–60)	100% (12/12)	N/A	N/A	N/A	N/A	Similar to sJCD VV1	JCD	Brown et al. (1992); Goldfarb et al., (1992); Kong et al. (2004)
D178N-129Va	Cis V	8		45 ± 6 (37–52)	(18–21) months	FHx score 2 (n = 4)	50% (1/2)	0% (0/1)	0% (0/2)	100% (2/2)			Takada et al. (2017)
D178N-129M	Cis M	106	Insomnia Sympathetic overactivity	Mean 52 (~20–76)	Median 12.6 (4–40) months	60–88%	14.4%	8%	2.1%	16.5%	Deg of Th (md and av nu) and inf olivary nu, little to no Vor PrPSc dep	FFI	Reder et al. (1995); Collins et al. (2001); Kovacs et al. (2005); Zarranz et al. (2005); Sano et al. (2013); Krasnianski et al. (2016)
V180I	Cis M	225	JCD phenotype, but w/ slower prog	Mean 77	Mean 25 months	0.7–6%	70%;	78.5%	11%	99%	V PrP-S	JCD	Kong et al. (2004); Higuma et al. (2013)
V180I*	Cis M	2		(84)	(21) months	FHx score 1 (n = 1) 2 (n = 1)	0% (0/1)		0% (0/1)	100% (1/1)			Takada et al. (2017)
T183A	Cis M	3	bvFTD, AD	45 ± 4 (42–49)	4 ± 2 years (2–9)	100% (2/ 2)	N/A	N/A	0% (0/7)	0% (0/2)	V & NL (CLs IV, V, VI), PrP (Cb, Pu) Small Plq-like PrP, Predom monoglycos PrPSC	JCD	Nitrini et al. (1997); Grasbon-Frodl et al. (2004)
H187R	MM/MV/VV	7	Early Cog and behavioral Sxs w/late Cb Atx Early Cb Atx and D after a few years Case w/ Pscyh Sxs in adolescence	(20–53)	(3–19) years	100% (4/4)	0% (0/2)		0% (0/4)	0% (0/4)	G multicentric PrP-Plqs in some cases. Curly PrP-G	GSS	Cervenakova et al. (1999); Butefisch et al. (2000); Hall et al. (2005); Colucci et al. (2006)
H187R*	Cis M	4		(30–41)	(12–13) years	FHx score 2 (n = 1)	0% (0/1)	0% (0/1)	0% (0/1)	0% (0/1)			Takada et al. (2017)
T188R	MV/VVV (cis V)	12	JCD	(55–66)	(14–16) months	0% (0/1)e	50% (1/2)	100% (1/1)	50% (1/2)	50% (1/2)	V, NL, A PrP-S and plaque-like PrP, type 1 PrPSc	JCD	Roeber et al. (2008); Tartaglia et al.(2010)
T188K	MM/MV (cis M)	3	JCD	Median 58 (39–76)	(2–13) months	8–37%e	69%		12%	69%	SE PrP-S	JCD	Roeber et al. (2008); Chen et al. (2013); Shi et al. (2015)
T188A	MM	1	JCD	(82)	(4) months	0% (0/1)	100% (1/1)	100% (1/1)	100% (1/1)	0% (0/1)	Sev G, V, Mod NL (predom in OLs) PrP-Neg	JCD	Collins et al. (2001)
T193I	MM		JCD	(70)	(10) months	0% (0/1)	100% (1/1)	100% (1/1)	100% (1/1)	0% (0/1)	N/A	N/A	Kotta et al. (2006)
E196K	N/A	13	JCD	(64–69)	(10–13) months	100% (1/1)	N/A	N/A	0% (0/1)	N/A	N/A	N/A	Peoc’h et al. (2000)
F198S	MV/VV	5	Cb Atx and D freq Park	(40–71) years	Mean 5 years (2–12)	100% (3/3)	N/A	N/A	N/A	0% (0/1)	Uni- and multicentric PrP-Plqs	GSS	Farlow et al. (1989); Dlouhy et al. (1992); Ghetti et al. (1995); Kong et al. (2004)
F198S*	Cis V	5		55 ± 8 (46–66)	67 ± 23 months (34–84)	FHx score 1 (n = 2) 2 (n = 1)	33.3% (1/3)	100% (1/1)	0% (0/4)	0% (0/3)			Takada et al. (2017)
F198V	MM	1	D w/ visual hallucinations, myoclonus and Park Clinical dx of early-onset AD	(56)	(4) years	N/A	N/A	N/A	0% (0/1)	0% (0/1)	N/A	N/A	Zheng et al. (2008)
E200K	MM/MV/VV	571	Similar to sJCD Peripheral neuropathy and supranuclear gaze palsy in some	Mean 60 (33–84)	(1–18) months	50%	85–100%	80–100%	42–85%	50–88%	Usually sJCD MM1 PrPSc types 1 and 2	JCD	Spudich et al. (1995); Meiner et al. (1997); Kovacs et al., (2005, 2011); Krasnianski et al. (2016)
E200K*	Cis M (n = 16) and cis V (n = 1)	34		60 ± 13 (36–84)	11 ± 17 (1–78)	FHx score 0 (n = 2) 1 (n = 10) 3 (n = 12)	57.1% (4/7)		37.5% (3/8)	88.9% (16/18)			Takada et al. (2017)
E200G	MV (cis V)	1	RPD, Cb Atx, Park ↓sensation in LEs	(57)	(30) months	0% (0/1)	0% (0/1)	100% (1/1)	0% (0/1)	100% (1/1)	SE w/ type 2 PrPSc	JCD	Kim et al. (2013)
D202G	MV (cis V)	1	Slowly progressive D w/ Cb Atx Later Pyram and EP signs	(55)	(16) years	100% (1/1)	100% (1/1)	0% (0/1)	0% (0/1)	0% (0/1)	N/A	N/A	Heinemann et al. (2008)
D202N	VV	1	D (AD) w/ Cb Atx	(73)	(6) years	N/A	N/A	N/A	N/A	N/A	PrP-Plqs, NFT	GSS	Piccardo et al. (1998)
V203I	N/A	17	JCD	(69)	(1) month	0% (0/1)	N/A	N/A	100% (1/1)	N/A	N/A	N/A	Peoc’h (2000)
R208H	MM/VV	15	D w/ behav changes Park and Pyram signs freq Report of a PSP-like phenotype	(58–63)	(3–16) months	20% (1/5)	50% (4/8)		57.1% (4/7)	25% (2/8)	SE, PrP-S (perineuronal perivacuolar) type 1 PrP	JCD	Capellari et al. (2005); Roeber et al. (2005); Matej et al. (2012); Vita et al. (2013); Shi et al. (2015)
V210I	Cis M	247	JCD	Mean 59 (39–82)	Median 5 (2–20) months	12–31%	90–100%	100%	44–80%	15–33%	Similar to sJCD MM1	JCD	Kong et al. (2004);Kovacs et al. (2005);Breithaupt et al. (2013);Krasnianski et al. (2016)
V210I*	Cis M	3		57 ± 15 (47–74)	(1) month	FHx score 0 (n = 2) 2 (n = 1)	100% (1/1)		33.3% (1/3)	100% (2/2)			Takada et al. (2017)
E211Q	MM	11	JCD	(42–81)	(6–32) months	100% (2/2)	N/A	N/A	100% (4/4)	N/A	V, G Mi PrP-S types 1 and 2 PrPSc	JCD	Peoc’h et al. (2000, 2012); Ladogana et al. (2001)
E211D	VV	1	Cb Atx, and late D	(53–68)	(3 –13) years	50% (1/2)	N/A	N/A	0% (0/2)	0% (0/2)		Multicentric PrP-Plqs Dystrophic neurites and NFT	GSS	Peoc’h et al. (2000, 2012)
Q212P	MM	2	Cb Atx w/o D Dx of olivoponto Cb degeneration	(60)	(8) years	N/A	N/A	N/A	N/A	N/A	Mod PrP Mi, PrP-Plqs	GSS	Piccardo et al. (1998)
I215V	MM	1	JCD	(55–76)	(12–15) months	0% (0/2)	(1/3)		100% (2/2)	50% (1/2)	NL, G, V, PrP-Neg	JCD
Q217R	VV/MV (cis V)	3	D w/ Cb Atx Cog decline, stereotypical behav Late Park and apraxia. Clinical dx of bvFTD and CBS	(45–66)	(5–13) years	100% (3/3)	N/A	N/A	0% (0/1)	0% (0/1)	Uni- and multicentric PrP-Plqs, NFT (neocortex)	GSS	Hsiao et al. (1992); Piccardo et al. (1998); Woulfe et al. (2005); Munoz-Nieto et al. (2013)
Y218N	VV	1	Atypic D w/AD and (54–61) bvFTD features Early language and executive impairment No Atx	(6) years	100% (1/1)	N/A	N/A	0% (0/2)	0% (0/2)	Uni- and multicentric PrP-Plqs, NFT w/ hyperP tau	GSS	Alzualde et al. (2010)
A224V	VV (cis V)	1	RPD	(48)	(32) months	0% (0/1) e		100% (1/1)	N/A	100% (1/1)	Diff V w/ PrPSc type 1	JCD	Watts et al. (2015)
M232R	MM	63	Similar to sJCD, some w/ slower prog	Mean 64 (15–81)	Mean 8 (0–32) months	~0%	55–75%	55–93%	20–100%	85%	sJCD MM1	JCD	Shiga et al. (2007); Zheng et al. (2008); Nozaki et al. (2010); Higuma et al. (2013)
M232T	MV		Cb Atx, spastic paraparesis and D	N/A	(6) years	0% (0/1)	N/A	N/A	N/A	N/A	Multicentric PrP-Plqs	GSS	Bratosiewicz et al. (2000)
P238S	N/A		JCD	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	N/A	Windl et al. (1999)

^aIncluding D178N-129V and D178N-129M.

^bBy nine prion disease (PrD) surveillance centers, according to Minikel et al. (2016).

^cPositive family history of dementia with similar clinical features (as of the proband) or PrD. For UCSF FHx (family history) score scale: 0 when there was no positive family medical history suspicious for or known PrD; 1 when there was at least one first-degree relative with dementia, encephalopathy, or movement disorder; or 2 in patients who were part of families with known PRNP mutations, or had positive history for clinical or path-proven PrDs.

^dAccording to most commonly used European 2009 and UCSF 2011 criteria (Zerr et al., 2009; Vitali et al., 2011).

^eThere is evidence of incomplete penetrance, as asymptomatic older carriers also were identified.

^fData on age at onset and duration of disease are shown as mean ± sd (range), unless otherwise indicated.

^gpositive if > total tau 1200 pg/mL.

If there are differences between data published in the literature from the more recently published University of California, San Francisco (UCSF) cohort, this information is provided in the table separately for that mutation.

Ab, abundant; AD, Alzheimer-type dementia; AH, Ammon horn; atypic, atypical; Atx, ataxia; av, anteroventral; BG, basal ganglia; bvFTD, behavioral variant frontotemporal dementia; Cb, cerebellum; CBS, corticobasal syndrome; CLs, cortical layers; Cog, cognitive; CSF, cerebrospinal fluid; c/w, consistent with; D, dementia; deg, degeneration; dep, deposition; diff, diffuse; dx, diagnosis; EEG, electroencephalogram; EP, extrapyramidal; ERC, entorhinal cortex; FFI, fatal familial insomnia; FHx, family history; F, focal; freq, frequent; G, gliosis; GSS, Gerstmann–Sträussler–Scheinker; GTC, generalized tonic-clonic seizures; HP, hippocampus; HyperP, hyperphosphorylated; inf, inferior; JCD, Jakob–Creutzfeldt disease; LE, lower-extremities; LMN, lower motor neuron; md, mediodorsal; Mi, mild; Min, minimal; Mod, moderate; Mol, molecular layer of the cerebellum; monoglycos, monoglycosylated; MRI, magnetic resonance imaging; N/A, not available; NFT, neurofibrillary tangles; NL, neuronal loss; nu, nuclei; OLs, occipital lobes; Park, parkinsonism; predom, predominant; prog, progression; PrP-G, granular PrP deposits; PrP-Neg, negative PrP staining; PrP-Plqs, PrP-amyloid plaques; PrP-S, synaptic PrP deposits; PSP, progressive supranuclear palsy; PSWC, periodic sharp-wave complexes; Pu, putamen; Pyram, pyramidal; Psych, psychiatric; RPD, rapidly progressive dementia; SE, spongiform (vacuolated) encephalopathy; Sev, severe; sJCD, sporadic Jakob–Creutzfeldt disease; SYN, syndrome; Th, thalamus; Sxs, symptoms; V, vacuolation; w/, with; WM, white matter; w/o, without.

PHENOTYPES ASSOCIATED WITH PRNP MUTATIONS

Genetic JCD (also called familial JCD)

More than 23 missense variants in PRNP have been known to cause gJCD (P105T, G114V, R148H, D178N (with codon 129 cis V), V180I, T183A, T188A, T188K, T188R, T193I, K194E, E196A, E196K, E200K, E200G, V203I, R208H, V210I, E211Q, I215V, A224V, M232R, and P238S) (Minikel et al., 2016; Takada et al., 2017). Clinical manifestations of these mutations are summarized in Table 29.1. Per a large European study, the mean age at onset in gJCD is approximately 60 years with great variability, from the second to ninth decades (Kovacs et al., 2005). Although median disease duration is about 5 months, at least 75% of patients die in less than 20 months and total durations of 8 years or more also have been reported (Kovacs et al., 2002). The mean age at onset for gJCD is reported to be about 7 years younger than for sJCD, but although disease duration is highly variable, it is similar to sJCD (a median of 4–7 months) (Brown et al., 1986; Parchi et al., 1999; Kovacs et al., 2005). Importantly, a large European study showed that 47% of PRNP mutation carriers diagnosed with gPrD did not have a known family history of PrD or other neurologic disorder. A more detailed investigation of the family history sometimes reveals dementia or neuropsychiatric illness that was likely misdiagnosed (Goldman et al., 2004; Kovacs et al., 2005). Negative family histories also might be due to incomplete penetrance, early death, de novo mutations, or nonpaternity (Mitrova and Belay, 2002; Kovacs et al., 2005).

Dementia is very common, occurring in approximately 95–98% of patients with gJCD (Meiner et al., 1997; Kovacs et al., 2002). Cerebellar symptoms (70%), myoclonus (60–70%) (Meiner et al., 1997; Kovacs et al., 2002), extrapyramidal signs (50%), and psychiatric symptoms (25%) are also frequently reported (Kovacs et al., 2002). Ancillary testing (cerebrospinal fluid (CSF) markers, electroencephalogram (EEG), and brain magnetic resonance imaging (MRI)) in gJCD showed lower diagnostic accuracies than in sJCD (Kovacs et al., 2002, 2005). Below we discuss two of the more common sJCD mutations worldwide, E200K and D178N; other PRNP missense mutations are summarized in Table 29.1.

The most common worldwide PRNP mutation for all gPrDs, including gJCD, is the E200K mutation (Lee et al., 1999; Kovacs et al., 2005; Minikel et al., 2016). It has at least four ancestral origins of patient clusters, the two largest among Sephardic Jews and Slovakians (Meiner et al., 1997; Lee et al., 1999; Mitrova and Belay, 2002). Penetrance of E200K is incomplete but still relatively high: 70% at age 70 and almost 100% at age 85 in Sephardic E200K carriers but slightly lower, about 60%, among Slovakians (Spudich et al., 1995; Mitrova and Belay, 2002). Interestingly, about 50% of E200K patients have negative family history for PrD (Spudich et al., 1995; Kovacs et al., 2005; Higuma et al., 2013), which might be due to several factors, including in our experience, that the disease is often hidden between branches of the family.

Most clinical manifestations of E200K are similar to sJCD (Meiner et al., 1997; Kovacs et al., 2005). The mean age at onset of symptoms is about 60 years (range of 33–84 years), a few years younger than sJCD (Meiner et al., 1997; Mitrova and Belay, 2002; Kovacs et al., 2005; Begue et al., 2011; Breithaupt et al., 2013; Higuma et al., 2013; Sano et al., 2013; Krasnianski et al., 2016; Takada et al., 2017), but the median disease duration is similar to sJCD, at about 5–6 months (Kovacs et al., 2005; Begue et al., 2011; Breithaupt et al., 2013; Krasnianski et al., 2016; Takada et al., 2017). Based on data from several national cohorts, dementia develops in ~95% of patients, cerebellar ataxia in ~80%, myoclonus in ~74%, pyramidal signs in ~61% (Meiner et al., 1997; Kovacs et al., 2011; Higuma et al., 2013; Krasnianski et al., 2016), and psychiatric symptoms such as hallucinations, depression, delusions, and aggressiveness in ~58% (Kovacs et al., 2011; Krasnianski et al., 2016). Less commonly, extrapyramidal signs have been reported in ~41% of patients (Meiner et al., 1997; Kovacs et al., 2011; Krasnianski et al., 2016), chorea/dystonia in ~33% (Kovacs et al., 2011; Krasnianski et al., 2016), and insomnia in ~26% (Meiner et al., 1997; Kovacs et al., 2011; Krasnianski et al., 2016). Certain symptoms appear to be more common in E200K than in sJCD, including seizures, headache, supranuclear gaze palsy, and peripheral neuropathy (Bertoni et al., 1992; Neufeld et al., 1992; Meiner et al., 1997; Kovacs et al., 2011; Krasnianski et al., 2016).

Regarding ancillary diagnostic biomarkers, the sensitivity of CSF total tau appears to be higher than CSF 14-3-3 in E200K patients (88.5% vs. 73%) (Kovacs et al., 2005, 2011; Breithaupt et al., 2013; Higuma et al., 2013; Sano et al., 2013; Krasnianski et al., 2016). The recently developed, highly specific ( ~98%) CSF reverse templating quake-induced conversion assay (RT-QuIC), has an average sensitivity of 83% in Japanese E200K subjects (Higuma et al., 2013; Sano et al., 2013). MRI changes typical for JCD (deep nuclei hyperintensity and cortical hyperintensities on diffusion-weighted imaging (DWI) or fluid-attenuated inversion recovery (FLAIR) MRI) (Fig. 29.2) are reported in at least 84–89% of E200K subjects. Involvement of the deep nuclei on T2, FLAIR, or DWI sequences appears to be more common than cortical ribboning (Breithaupt et al., 2013; Takada et al., 2017), which is the reverse of what is reported to be found in sJCD (Vitali et al., 2011). Periodic sharp-wave complexes (PSWCs) on EEG have been reported only in 55% of E200K cases (Kovacs et al., 2005, 2011; Breithaupt et al., 2013; Krasnianski et al., 2016), lower than the 66% in sJCD (Steinhoff et al., 2004). Neuropathology of E200K is very similar to the most common subtype ofof sJCD, except the neocortex, basal ganglia, and thalami are more severely affected in E200K (Kovacs et al., 2011; Higuma et al., 2013).

Fig. 29.2.

Brain magnetic resonance imaging (MRI) in genetic prion disease with E200K PRNP mutation. (A) Diffusion-weighted imaging (DWI) MRI of a 47-year-old E200K patient with genetic Jakob–Creutzfeldt disease (gJCD) 3 months after onset shows diffuse cortical hyperintensity (cortical ribboning; dashed arrows) of the bilateral frontal and insula (left > right) and temporo-parietal cortices. There was also DWI hyperintensity in the bilateral striata (left > right, solid arrows). (B) The ADC map of the same gJCD E200K case showed hypointensity in most of the regions that were hyperintense on DWI, confirming reduced diffusion, consistent with prion disease. Orientation is radiologic. (Reproduced from Takada et al. (2017), with permission from John Wiley.)

That the same mutation, D178N, could lead to two different phenotypes (gJCD or FFI) was puzzling at first, but Goldfarb et al. (1992) showed that the PRNP codon 129 polymorphism found on the cis (mutated) allele had a strong effect on the phenotype; FFI was associated with D178N and methionine at codon 129 (129M) and D178N gJCD was usually assocatied with valine at codon 129. Data since then suggest the effect is not absolute and that D178N presents along a spectrum between gJCD and FFI (Zerr et al., 1998; Zarranz et al., 2005). Aside from influencing the clinicopathologic phenotype of gPrD, the PRNP codon 129 polymorphism is an important genetic risk factor for sporadic and acquired PrD, as discussed above.

The mean age at onset of JCD with D178N-129V is about 46 years, ranging from the second to the eighth decades (Brown et al., 1992; Goldfarb et al., 1992), and the mean disease duration is about 23 months (range 7–60 months) (Brown et al., 1992; Goldfarb et al., 1992). Initial symptoms include cognitive decline (particularly memory, 95% of cases), behavioral changes (30%), and cerebellar symptoms (21%). During the clinical course, dementia (about 100%), cerebellar symptoms (86%), myoclonus (74%), extrapyramidal symptoms (67%), pyramidal symptoms and signs (50%), and seizures (12%) were reported to develop (Brown et al., 1992). EEG PSWCs are reported in few (Brown et al., 1992; Nozaki et al., 2010), and to our knowledge, no large cohort data on MRI and CSF findings in D178N-129V cases exist, however. The neuropathology of D178N-129V is similar to that of a subtype of sJCD (VV1); the neocortex exhibits more severe spongiosis and neuronal loss but they are very mild in the thalamus and absent in the cerebellum (Gambetti et al., 2003). A summary of the clinical phenotypes with other mutations causing gJCD is provided in Table 29.1.

Gerstmann–Sträussler–Scheinker disease

GSS disease is defined as a chronic hereditary autosomal-dominant prionopathy with multifocal amyloid plaque deposition composed of PrP (PrP-amyloid plaques) in the neuropathologic exam (Ghetti et al., 1995). It was the first human PrD in which a PRNP mutation was identified, and since its original description in an Austrian family in 1936, a wide variety of clinical phenotypes have been described (Masters et al., 1981; Hsiao et al., 1989). The presence of multicentric PrP-amyloid plaques in the neuropathologic exam remains the key feature of the disease, differentiating GSS from most of the other gPrDs (Kong et al., 2004; Liberski, 2012).

To date, more than 20 missense and OPRI mutations causing GSS have been described worldwide (Tables 29.1 and 29.2). The prevalence of GSS disease is difficult to ascertain (it has been estimated in a range of 1–10/100,000,000) since it is thought to be largely unrecognized (Kong et al., 2004). Several reasons may account for the elusiveness of the diagnosis: first of all, its clinical heterogeneity, with potential similarity to sJCD and other neurodegenerative diseases such as Alzheimer disease, frontotemporal dementia, multisystem atrophy, Parkinson disease, and amyotrophic lateral sclerosis (Kong et al., 2004). In addition, up to 30% of GSS cases show no evidence of family history, making the diagnosis challenging (Kovacs et al., 2005).

Table 29.2.

Octapeptide repeat insertions and deletions

PRNP mutation	Codon 129 polymorphism	# of cases in literature	Clinical phenotypes	Age at onset (range)c (years)	Disease duration (range)c (months or years)	Positive FHxa	CSF marker sensitivity		EEG PSWC	MRI c/w JCDb	Neuropathology	Neuropathology pheno-type	References
							14-3-3	Total taud
2-OPRD	MM (1) Unknown (1)	2	RPD RPD, Sz, Myo	(62–86)	(18–23) months	0% (0/2)	N/A	N/A	N/A	N/A		JCD	Beck et al., (2001); Capellari et al. (2002)
2-OPRI	MM (1) MV (1) VV (1) Unknown (1)	4	RPD D Cbr Atx	Mean 63.3 ± 7.9 (58–75)	Mean 6.8 ± 6.4 (0.25–13) years	50% (2/4)	N/A	N/A	N/A	0% (0/1)		JCD	Goldfarbe et al. (1993); Croes et al. (2004); van Harten et al.(2000)
3-OPRI	VV (1) MM(1)	2	RPD	(68–69)	(4 months–3 years)	0% (0/1)	50% (1/2)	N/A	50% (1/2)	100% (1/1)		JCD	Grasbon-Frodl et al. (2004) ; Nishida et al.(2004)
4-OPRI	MM(9) VV (1) MV (Cis V, 1)	11	MM: RPD, Myo, Cbr, Atx VV: Dep, Behav MV: RPD	MM: Mean 60 ± 13.6 (39–85) VV: 82 MV: 38	MM: mean 33.8 ± 32.7 (2–76) months VV: 4 months MV 8 months	MM: 12.5% (1/8) VV: 0% (0/1) MV: 100% (1/1)	MM: 100% (4/4) VV: N/A MV: 100% (1/1)	N/A	MM:22% (2/9) VV: 100% (1/1) MV: 0% (0/1)	MM: 20%(1/5) VV: N/A MV: 100% (1/1)		MM: JCD (7/7) VV: N/A MV: JCD (1/1)	Laplanche et al. (1995); Kaski et al. (2011); Sanchez-Valle et al. (2012)
5-OPRI	MM (6) MV (Cis M, 3) Unknown (8)	17	Cog, motor	Mean 45.7±11.0 (26–63),	Mean 76 ± 51.8 (10months–14.5 years)	100% (15/15)	100% (1/1)	N/A	25%(2/8)	0% (0/8)	V, Neu loss, kuru-like PrPSc		Goldfarb et al. (1991); Cochran et al. (1996); Skworc et al. (1999); Beck et al. (2005); Mead et al. (2007)
5-OPRI*	MM (1)	1	Visuosp	39	19Y	FHx score 2	N/A	N/A	0%	0%	N/A		Takada et al. (2017)
6-OPRI		63	Cog D, Front, Cbr Atx	MM (30) 31.4 MV (10) 41.7	MM (19) 11.4Y, MV (8) 8.9Y		N/A	N/A	0%	0%			Mead et al. (2006)
6-OPRI*	Cis M (3) Cis V (2)	5	Cog, Cbr Atx	Cis M 35±2.6 (32–37) Cis V (47–51)	Cis M 5.9±3(3–9) years Cis V (5–10) months	FHx score 2(n = 5)	0% (0/2)	0% (0/2)	0% (0/2)	0% (0/5)		JCD (4/4)	Takada et al. (2017)
7-OPRI	Cis M/Cis V	16	Cog, Behav Motor	35 ±12.4 (18–59)	8.4 ±4.9, (0.6–17)	86% (6/7)	N/A	N/A	33.3% (1/3)	50% (1/2)	Cis M no PrP-Plqs Cis V Uni-, multicentric PrP-Plqs	Goldfarb et al. (1991); Tateishi et al. (1991); Brown et al. (1992); Dermaut et al. (2000) Lewis et al. (2003); Cannella et al. (2007) Wang et al. (2007); Guo et al. (2008); Mauro et al. (2008); Jansen et al. (2011)
8-OPRI	Cis M (4)	11	Psy, D	Mean 28 (21–34)	Mean 3.8 (1–7) years	100% (11/11)	N/A	N/A	0% (0/3)	0% (0/2)	Kuru, multicentric PrP-Plqs	GSS	Laplanche et al. (1999)
8-OPRI*	MM	1	Dep, Cbr Atx	22	>5 yearse	FHx score 0	N/A	N/A	N/A	0%	N/A		Takada et al. (2017)
9-OPRI	Cis M (2) Unknown (1)	3	Cbr Atx Fall, Cog Behav, Cog	47±13 (32–55)	(7 months–2.5 years)	N/A	N/A	N/A	0% (0/1)	0% (0/1)	Numerous small PrP plaqs		Owen et al. (1992); Duchen et al. (1993); Krasemann et al.(1995)
9-OPRI*	VV (1)	1	Visuosp, Cog	47	21 months	FHx score 0	N/A	100% (1/1)	100% (1/1)	100% (1/1)		JCD+ GSS	Takada et al. (2017)
12-OPRI	N/A	3	Cog, Behav, Cbr Atx, Sz	44 ± 1 (43–45)	8±1.7 (7–10)	100% (3/3)	0% (0/1)	100% (1/1)	0% (0/1)	0% (0/1)	Multicentric PrP-Plqs	GSS	Kumar et al. (2011)

^aPositive family history of dementia with similar clinical features (as of the proband) or prion disease (PrD). For UCSF family history (FHx) score scale: 0 when there was no positive family medical history suspicious for or known PrD; 1 when there was at least one first-degree relative with dementia, encephalopathy, or movement disorder; or 2 in patients who were part of families with known PRNP mutations, or had positive history for clinical or path-proven PrDs.

^bAccording to most commonly used European 2009 and UCSF 2011 criteria (Zerr et al., 2009; Vitali et al., 2011).

^cData on age at onset and duration of disease are shown as mean ± sd (range), unless otherwise indicated.

^dPositive if > total tau 1200 pg/mL.

^eSome patients still alive, so duration at last follow-up.

If there are differences between data published in the literature from the more recently published University of California, San Francisco (UCSF) cohort, this information is provided in the table separately for that mutation. Atx, ataxia; Cbr, cerebellar; c/w, consistent with; Behav, behavioral changes; Cog, cognitive; CSF, cerebrospinal fluid; D, dementia; Dep, depression; EEG, electroencephalogram; FHx, family history; Front, frontal-lobe dysfunction; GSS, Gerstmann–Sträussler–Scheinker; JCD, Jakob–Creutzfeldt disease; MRI, magnetic resonance imaging; Myo, myoclonus; N/A, not available; Neu, neuronal; OPRD, octapeptide repeat deletion; OPRI, octapeptide repeat insertion mutation; PrP-Plqs, PrP-amyloid plaques; PSWC, periodic sharp-wave complexes; Psy, psychiatric changes; RPD, rapidly progressive dementia; Sz, seizures; Visuosp, visuospatial.

A proline-to-leucine mutation at codon 102 in PRNP (P102L) is the most common mutation causing GSS worldwide. Typically, symptoms begin in the fifth decade (with a range of age at onset from 27 to 66 years), with illness durations ranging from 7 months to 11 years and a mean duration of 4 years. A slowly progressive cerebellar syndrome was the commonest clinical presentation and cerebellar signs are almost universal during disease progression. There is a subset of patients with prominent psychiatric onset (and prominent frontal executive on neuropsychologic testing) with early onset and death. Even in the same family, presentations can differ (Hainfellner et al., 1995; Popova et al., 2012). Cognitive symptoms, prominent with disease progression, are infrequently reported as the first manifestation of the disease; however, these may be under-recognized because of the lack of neuropsychologic assessment. Lower motor neuron signs with areflexia, often associated with objective evidence of muscle weakness, myopathic gait, and extrapyramidal signs (parkinsonism), have been reported in more than half of cases. In addition, sensory symptoms such as dysesthesia are increasingly recognized and may reflect an associated sensorimotor axonal neuropathy. Conversely, myoclonus, dystonia, and apraxia only are reported occasionally. Irrespective of the clinical presentation, there is usually a gradual progression of cerebellar and pyramidal dysfunction with behavioral symptoms and cognitive decline leading to a bed-ridden akinetic-mutism state (Kong et al., 2004; Webb et al., 2008). GSS-associated missense mutations with their corresponding phenotype are summarized in Table 29.1 (GSS-associated OPRI mutations are discussed below and in Table 29.2).

Additionally, nonsense mutations of PRNP leading to PrP truncation are associated with unusual phenotypes, sometimes classified as GSS-like, with prolonged clinical courses, severe neurofibrillary tangle pathology, and high levels of cerebral amyloidosis (Mead et al., 2013; Guerreiro et al., 2014).

The MRI is less diagnostically helpful in GSS than in sJCD, usually showing general cortical and/or cerebellar atrophy, although early MRI can be normal (Arata et al., 2006). Diffusion abnormalities usually are not found in GSS (Vitali et al., 2011). Positron emission tomography (PET) ligands binding nonspecific protein-amyloid deposits may offer valuable diagnostic information and could be used in the future as a treatment-monitoring tool. In fact, [F-18] FDDNP PET ligand has been reported to identify plaques in both symptomatic and asymptomatic patients with GSS (P102L, A117V, F198S) (Kepe et al., 2010). EEG does not show the typical JCD findings of PSWCs and CSF biomarker proteins such as 14-3-3 and total tau are elevated less often than in sJCD, perhaps in less than half of cases (Kovacs et al., 2005; Jones et al., 2014).

Familial fatal insomnia

FFI is about the third most common gPrD PrD worldwide and has a relatively unique clinicopathologic phenotype. It is caused by aspartate (D) being substituted for asparagine (N) at PRNP codon 178 coupled with cis methionine at codon 129 (D178N, 129M) (Lugaresi et al., 1986; Goldfarb et al., 1992; Gambetti et al., 2003). In FFI, not only does the cis codon 129 allele have a strong effect on the phenotype, but the trans codon 129 allele (MM vs. MV) may affect clinical feature, age of onset, and disease duration (Krasnianski et al., 2008, 2014). Even within a family there can be great variability in presentation, some presenting as FFI, others as gJCD (Synofzik et al., 2009).

The age at presentation for D178N 129M varies from 20to 70 years (median 55–60), with illness duration averaging around 13–15 months (Reder et al., 1995; Collins et al., 2001; Krasnianski et al., 2014); codon 129 homozygotes (129MM) have shorter duration but later onset than 129MV (Gambetti et al., 2003; Krasnianski et al., 2014). Incomplete penetrance has also been recognized in FFI. The core clinical features of FFI are linked to thalamic dysfunction, and consist of profound disruption of the normal sleep–wake cycle (with complete disorganization of the EEG patterns of sleep), sympathetic overactivity, diverse endocrine abnormalities (with attenuation of the normal circadian oscillations), and markedly impaired attention. Neuropathology typically shows a characteristic restricted degeneration of the thalami with severe neuronal loss gliosis, particularly in the mediodorsal and anteroventral nuclei, as well as the inferior olivary nuclei and isolated gliosis in the midbrain and hypothalamic gray matter. Deposition of PrP^Sc is present in these same regions, whereas vacuolation (spongiform change) typical of most other PrDs is usually absent. Longer-duration cases have involvement of the cerebral cortex (Lugaresi et al., 1986; Medori et al., 1992; Gambetti et al., 2003).

Clinical variability is seen across mutation carriers, some presenting with moderate insomnia and lacking abnormal EEG sleep studies, even despite showing the characteristic thalamic gliosis at neuropathology (Zerr et al., 1998; Taniwaki et al., 2000; Collins et al., 2001). Signs and symptoms of autonomic dysfunction were thought be one of the earlier features of the disease; however, psychiatric symptoms and abnormal gait have also been identified as common presenting symptoms in D178N 129M carriers (Zarranz et al., 2005). Moreover, nonspecific symptoms such as marked weight loss and tiredness can be the first manifestations of the disease, making the diagnosis challenging (Johnson and Gibbs, 1998). A positive family history is only found in about two-thirds of FFI cases (Krasnianski et al., 2014).

In general, ancillary tests, such as CSF, EEG, and MRI, are not helpful for FFI diagnosis. The clinical diagnosis of FFI is solely based on the careful observation of the clinical course, and only polysomnography and, to some extent, PET scan showing thalamic gliosis or hypometabolism (Cortelli et al., 1997, 2006; Haik et al., 2008) may contribute to the FFI diagnosis before genetic testing identifies a PRNP mutation . In fact, abnormal sleep patterns can be missed without a polysomnography (showing rapid eye movement and deep sleep reduction) (Zarranz et al., 2005; Krasnianski et al., 2008). The phenotypic heterogeneity and dearth of good ancillary diagnostic tests can make diagnosis challenging. A new diagnostic algorithm with the objective of improving early recognition of FFI patients and allowing diagnosis in the absence of genetic testing has recently been proposed (Krasnianski et al., 2014).

Atypical presentations of gPrD

gPrD due to PRNP octapeptide repeat insertions

gPrD due to OPRIs are highly variable, both in clinical and neuropathologic characteristics. OPRIs with up to four repeats tend to resemble sJCD clinically, whereas those patients with five to seven repeats usually are associated with the JCD phenotype, but with earlier onset and longer disease duration than typical sJCD. Patients with eight or more repeats often are reported to have a GSS phenotype, with widespread PrP^Sc amyloid deposition, including plaques (Gambetti et al., 2003; Takada et al., 2017). This phenotypic breakdown by size of the OPRI is not always accurate, however (Takada et al., 2017). Brain MRI usually shows diffuse cortical atrophy without typical JCD MRI findings of restricted diffusion (Takada et al., 2017). A more detailed description of OPRI mutations is provided in Table 29.2.

gPrD due to nonsense PRNP mutations

Nonsense mutations are very rare and have only been reported in a few kindreds. The clinical presentation is variable, but overall they are characterized by features very atypical for PrD. Some features can include prolonged disease courses (few years to more than a decade), clinical Alzheimer’s disease phenotypes, sensory and autonomic peripheral nervous system involvement, chronic gastrointestinal upset (sometimes cyclical), and presence of PrP amyloid plaques and/or PrP cerebral amyloid angiopathy, often combined with tau pathology in the brain (Mead et al., 2013; Guerreiro et al., 2014).

The Q160X mutation (with either cis methionine or valine at codon 129) has been reported in three families, with onset of symptoms ranging from 27 to 59 years of age (Finckh et al., 2000; Jayadev et al., 2011; Guerreiro et al., 2014; Fong et al., 2016). In two of those families, the probands were diagnosed clinically with Alzheimer disease, and in two individuals from one kindred the neuropathologic assessment revealed neurofibrillary tangles and amyloid plaques that immunostained for PrP, but not amyloid β. Chronic diarrhea was also reported two individuals. Other nonsense mutations, such as Q145X-129M, Q163X-129V, Y226X-129V, and a 2bp deletion at codon 178 (Ghetti et al., 1996; Jansen et al., 2010; Matsuzono et al., 2013; Mead et al., 2013), are summarized in Table 29.3.

Table 29.3.

PRNP nonsense mutations

PRNP mutation	Codon 129 polymorphism	# of cases in literature	Clinical phenotypes	Age at onset (range)c (years)	Disease duration (range)c (months or years)	Positive FHxa	CSF marker sensitivity		EEG PSWC	MRI c/w JCDb	Neuropathology	Neuropathology pheno-type	References
							14-3-3	Total tau
Q145X	MM	1	Cog	38	21 years	0% (0/1)	N/A	N/A	N/A	N/A	NFT, PrP-angio	Atypic	Ghetti et al. (1996)
Q160X	MM (4) MV (3) Unknown (4)	11	D Cog, Dep Cog, dysauto, neuropathy	42.1±8.4 (32–59)	9.6±4.9 (4–21) years	100% (11/11)	0% (0/1)	0% (0/1)	0% (0/4)	0% (0/4)	NFT, PrP-angio, AD pathology	Atypical	Owen et al. (1989); Finckh et al.(2000) ; Jayadev et al. (2011); Fong et al., 2016
Q163X	Cis V (10)	10	Dysauto, neuropathy, Cog	33.0 ±3.4 (30–38)	26.8 ±8.0 (15–33) years	100% (10/10)	100% (1/1)	100%d (1/1)	0% (0/3)	0% (0/2)	NFT, PrP-angio, PrP-Plaqs, Sp, V	Atypical	Mead et al. (2013)
Y226X	MV (2) Unknown (1)	3	D Park, Cog	55.3 ±17.0 (39–73)	3.5 ±2.3 (1.5–6) years	100% (3/3)	100% (1/1)	N/A	100% (1/1)	0% (0/1)	PrP-angio, PrP-Plaqs	Atypical	Jansen et al. (2010)
2bp Del 178	Unknown (3)	3	Dysauto, cog, neuropathy	42.0± 14.0 (26–52)	5.5± 6.4 (1–10) years	100% (3/3)	100% (2/2)	100% (1/1)	N/A	0% (0/2)	N/A	Atypical	Matsuzono et al. (2013)

From Kim et al. (2017) with permission from Cold Spring Harbor Laboratory Press.

^aPositive family history of dementia with similar clinical features (as of the proband) or prion disease.

^bAccording to most commonly used European 2009 and UCSF 2011 criteria (Zerr et al., 2009; Vitali et al., 2011).

^cData on age at onset and duration of disease are shown as mean ± sd (range), unless otherwise indicated.

^dPositive if > total tau 1200 pg/mL.

AD, Alzheimer-type dementia; Cog, cognitive; c/w, consistent with; D, dementia; Dep, depression; Dysauto, dysautonomia; EEG, electroencephalogram; FHx, family history; JCD, Jakob–Creutzfeldt disease; MRI, magnetic resonance imaging; N/A, not available; NFT, neurofibrillary tangles; Park, parkinsonism; PrP-angio, PrP amyloid angiopathy; PrP-Plqs, PrP-amyloid plaques; PSWC, periodic sharp-wave complexes; Sp, spongiosis; V, vacuolation.

SUMMARY AND FUTURE DIRECTIONS

Earlier ages at onset, longer disease durations, and lower ancillary testing positivity in gPrD compared with sJCD can make the diagnosis of gPrD difficult. Compounding this is lack of clear family history in many cases. Considering a gPrD on a diagnostic differential of an atypical neurologic syndrome can help diagnose gPrD.

Recent advances in diagnostic testing also can aid diagnosis of gPrD; however, one must be considering PrD in order to conduct these tests. The RT-QuIC assay that detects PrP^Sc in CSF and olfactory epithelia has shown 87–100% sensitivity and specificity in certain types of gPrD (Sano et al., 2013), but needs to be assessed in larger cohorts and more PRNP mutations. Furthermore, as genetic testing, dementia-associated gene panels, whole and even whole-genome exome sequencing become more available at a reduced cost, this should improve the diagnosis of genentic forms of dementia (Beck et al., 2014).

When considering PRNP genetic testing, it is recommended to follow the Huntington disease protocol for genetic testing, including providing genetic counseling (MacLeod et al., 2013). For predictive testing of at-risk relatives, psychiatric and neurologic evaluation should be performed prior to genetic testing, to better assess and help mitigate against the risks of testing, such as psychologic distress or depression. Genetic counselng can help persons at risk understand the positive and negative implications of testing (Goldman, 2015).

Unfortunately, all PrDs are incurable at present. Furthermore, we cannot tell who will develop the more common sporadic PrD, but for gPrDs we can identify carriers in the presymptomatic phase. Thus, when experimental or approved treatments become available they might be excellent candidates for preventive intervention.