Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
Mol Cell. Author manuscript; available in PMC 2012 January 21.
Published in final edited form as:
PMCID: PMC3049456

How one bad protein spoils the barrel: structural details of β2- microglobulin amyloidogenicity


In this issue, Eichner et al. (2011) describe at atomic resolution the structure of an amyloidogenic state of β2-microglobulin and how it may corrupt a soluble counterpart in the pathological scenario that ensues when good proteins go to the ‘dark side’ and form infectious toxic amyloid.

Evolutionary pressures select protein sequences that fold and function well enough for an organism to compete successfully and for the species to be preserved. Yet these evolutionary requirements create a precarious balance: Stochastic events and accumulated stresses can readily cause proteins to be unstable. Biology copes with this challenge by using robust systems of protein quality control, including molecular chaperones and degradation machinery. However, even these coping mechanisms are imperfect, particularly given the double challenges of longer life spans and increasing environmental insults. The result: numerous diseases directly attributable to protein misfolding, which leads to loss of function and, in many cases, toxic aggregates and protein fibrils known as amyloid. Amyloidogenic species may be toxic not only to the individual harboring them, but also to others by corrupting their otherwise healthy proteins. Two fundamental questions have remained unanswered because of the difficulty of studying transitory and ill-behaved amyloidogenic species: What are the structural features that underlie amyloidogenicity? And by what molecular mechanism do these species cause formerly stable, soluble proteins to become amyloidogenic? Clearly, answers to these questions will enhance the likelihood of successful therapeutic strategies against amyloid diseases. In this issue, Eichner et al. (2010) report an atomic-level structure of an amyloidogenic state of β2-microglobulin (β2m). Moreover, they demonstrate that this conformational state is capable of transforming soluble, well-folded β2m into an amyloidogenic species and postulate how this conversion takes place.

Arguably the best-studied amyloid disease-causing protein, β2m is normally an integral component in the Class 1 major histocompatibility complex (MHC-1), which resides on the surface of T-lymphocytes and other cells. In the normal course of events, the MHC-1 complex sheds β2m into the plasma, and the kidney filters and degrades it. Compromised kidneys are unable to perform this task, leading to an accumulation of 10- to 60-fold more β2m in the plasma than in a person with full renal capacity. The higher plasma concentration of β2m is directly correlated with the build up of amyloid deposits in kidney patients. As such patients are generally sustained by dialysis treatment, the amyloid disease caused by β2m is called dialysis-related amyloidosis (DRA). A majority of patients who are on dialysis for over 5 years will develop DRA.

The 99-residue long β2m adopts a canonical immunoglobulin fold, with seven β-strands (A to G) stabilized by a single disulfide bond from strand B to F (Fig. 1A). MHC-1 assembly in the endoplasmic reticulum requires association of folded β2m with the heavy chain along with the antigenic peptide to be presented. Folding of β2m is rate-limited by isomerization of the His31-Pro32 bond from trans to cis (Eichner and Radford, 2009; Jahn et al., 2006; Kameda et al., 2005). In vitro, the resulting long-lived intermediate (termed IT) is aggregation-prone and likely resembles the amyloidogenic state. This in vitro result has led to the general model that the amyloidogenic species in vivo likely contains a trans 31–32 peptide bond. However, it is unclear how plasma β2m that was presumably natively folded with a cis His31-Pro32 bond when shed from the MHC-1 complex might convert to a trans bond-containing amyloidogenic state.

Figure 1
(A) The high-resolution solution structures of human β2m (A) and its N-terminally truncated variant, ΔN6 (B), as reported by Eichner et al. (2011). [The lowest energy structures from NMR analysis are depicted in these PyMol images]. The ...

Provocatively, a number of perturbants cause natively folded, soluble β2m to become aggregation-prone in vitro, including low pH, treatment with Cu+2, and the action of detergents, co-solvents, or chaotropes, and each condition appears to promote an IT-like state (Platt and Radford, 2009). Intriguingly, a fraction of the β2m amyloid fibrils isolated from DRA patients is proteolytically clipped between residues 6 and 7 (termed ΔN6) (Bellotti et al., 1998), and in vitro studies showed that ΔN6 is strikingly more aggregation-prone than full-length β2m under conditions of physiological temperature and pH, and, like the full-length version, its aggregation is enhanced at acidic pH (Esposito et al., 2000). Crucially, by careful design of the experimental conditions ΔN6 proved amenable to the high-resolution structure determination reported in this issue by Eichner et al. (2011), revealing the first atomic resolution structure of this amyloidogenic state.

The detailed picture of the structural changes caused by removal of the six N-terminal amino acids from β2m is rich and hugely informative (Fig. 1B): First and foremost, the His31-Pro32 bond is trans in ΔN6, and a direct comparison of NMR signals shows compellingly that ΔN6 adopts an IT-like structure. But more stunning is the observation that 17 of 21 side chains packed in the hydrophobic core of β2m experience significant restructuring, with both Phe30 and Phe62 side chains moving more than 7 Å and ending up surface exposed in the amyloidogenic state. These rearrangements show how profound an effect the presence of the N-terminal six residues has on the metastable native fold of β2m: This region acts as a ‘lock’ in the native state, ensuring that the His31-Pro32 bond remains in the cis form, the geometry of which (Fig. 1C) in turn secures the native side chain packing and avoids populating an aggregation-prone state. Moreover, any perturbation to β2m that weakens interactions between the N-terminal segment and the rest of the molecule will have the same effect as truncation of this region and tip the balance towards the IT-like amyloidogenic state. Specifically, the structure now available for ΔN6, and thus IT, provides opportunity for detailed comparisons with the M* state proposed to be key to how Cu+2 enhances β2m amyloidogenicity (Calabrese and Miranker, 2009).

Protein aggregation generally follows nucleation-polymerization kinetics, displaying a lag phase that reflects the requirement for a low probability nucleus to form before subsequent favorable association events. In the case of β2m, the amyloidogenic state populated by the ΔN6 variant or by full-length protein upon destabilization by various treatments is a gateway to the rare nucleus that must form for aggregation to proceed. Eichner et al. (2011) provide evidence that enhanced conformational dynamics and subsequent protonation of His84 caused by mild acidification lead to greater access to a state with remodeling of loops BC and DE such that a dimer interface is created, and posit that this dimer represents a key stepping stone en route to the obligatory nucleus for an aggregation cascade.

One of the most exciting aspects of this story is the finding that the human ΔN6 variant can act in a prion-like manner, causing full-length human β2m (but not β2m from mouse) to form amyloid-like fibrils in weeks under conditions where otherwise it would be soluble for months to years. How does the Δ6N variant pass on its amyloidogenicity? A clever NMR experiment using ΔN6 monomers prepared with 14N enabled the authors to observe increased dynamics of full-length 15N-labeled natively-folded human β2m in the AB loop and partial fraying of the A strand from the rest of the β-sandwich upon addition of the ΔN6 variant. The resulting enhanced N-terminal dynamics, caused here by encounter with ΔN6 or in other cases by pH, mutation, or chaotropes, then triggers cis/trans isomerization of the His3-Pro32 peptide bond and conversion of the protein to the amyloidogenic state. Unraveling the subsequent structural evolution of the protein during later stages of amyloid formation awaits further experimentation, although progress is being made (for example, see Debelouchina et al., 2010). It will be exciting to see how the population of a rare conformer that has a strong tendency for intermolecular association then leads to the regular cross-β structure that is a hallmark of amyloid fibrils.

How do these results explain amyloidogenicity of β2m shed from cells? Presumably, environmental factors within the body trigger conversion to an amyloidogenic state by tickling the N-terminal region of this metastable protein. Following this logic, proteolytic cleavage of the N-terminal portion of plasma β2m would not only lead to increased amyloidogenicity of the clipped product but also create a species that can cause the conversion of fellow full-length soluble β2m species into amyloidogenic states (Fig. 1D). How, where, and when the cleavage of β2m to produce ΔN6 occurs physiologically must now be answered. If this cleavage is a major factor in DRA, then inhibition of β2m proteolysis emerges as a potential route towards therapy.

Why does evolution allow the risk of β2m amyloidosis to continue? There are two intriguing considerations: First, the metastability of this protein may confer an advantage at an earlier stage, here most likely in the formation of MHC-1 complexes. This may be due to the timing of trans-cis isomerization, for example, which may help coordinate folding of a multi-chain complex. This would be similar to the folding of immunoglobulins, which is also rate-limited by peptide bond isomerization of the light chains (Feige et al., 2009). The evolutionary selection for a metastable native state is reminiscent of serpins. They too are vulnerable to aggregation-associated pathologies but the risk is acceptable to preserve their capacity to perform wide-ranging and essential functions (Whisstock and Bottomley, 2006). Second, the risk of DRA from β2m arises only during kidney failure and thus is part of a more complex set of physiological breakdowns. It is difficult to imagine an evolutionary route to minimizing secondary pathological consequences in the presence of a primary organ failure.


Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.


  • Bellotti V, Stoppini M, Mangione P, Sunde M, Robinson C, Asti L, Brancaccio D, Ferri G. Beta2-microglobulin can be refolded into a native state from ex vivo amyloid fibrils. Eur J Biochem. 1998;258:61–67. [PubMed]
  • Calabrese MF, Miranker AD. Metal binding sheds light on mechanisms of amyloid assembly. Prion. 2009;3:1–4. [PMC free article] [PubMed]
  • Debelouchina GT, Platt GW, Bayro MJ, Radford SE, Griffin RG. Intermolecular Alignment in beta(2)-Microglobulin Amyloid Fibrils. J Am Chem Soc 2010 [PMC free article] [PubMed]
  • Eichner T, Radford SE. A generic mechanism of beta2-microglobulin amyloid assembly at neutral pH involving a specific proline switch. J Mol Biol. 2009;386:1312–1326. [PubMed]
  • Eichner T, Kalverda AP, Thompson GS, Homans SW, Radford SE. 2011 This issue. [PMC free article] [PubMed]
  • Esposito G, Michelutti R, Verdone G, Viglino P, Hernandez H, Robinson CV, Amoresano A, Dal Piaz F, Monti M, Pucci P, et al. Removal of the N-terminal hexapeptide from human beta2-microglobulin facilitates protein aggregation and fibril formation. Protein Sci. 2000;9:831–845. [PubMed]
  • Feige MJ, Groscurth S, Marcinowski M, Shimizu Y, Kessler H, Hendershot LM, Buchner J. An unfolded CH1 domain controls the assembly and secretion of IgG antibodies. Mol Cell. 2009;34:569–579. [PMC free article] [PubMed]
  • Jahn TR, Parker MJ, Homans SW, Radford SE. Amyloid formation under physiological conditions proceeds via a native-like folding intermediate. Nat Struct Mol Biol. 2006;13:195–201. [PubMed]
  • Kameda A, Hoshino M, Higurashi T, Takahashi S, Naiki H, Goto Y. Nuclear magnetic resonance characterization of the refolding intermediate of beta2-microglobulin trapped by non-native prolyl peptide bond. J Mol Biol. 2005;348:383–397. [PubMed]
  • Platt GW, Radford SE. Glimpses of the molecular mechanisms of beta2-microglobulin fibril formation in vitro: aggregation on a complex energy landscape. FEBS Lett. 2009;583:2623–2629. [PMC free article] [PubMed]
  • Whisstock JC, Bottomley SP. Molecular gymnastics: serpin structure, folding and misfolding. Curr Opin Struct Biol. 2006;16:761–768. [PubMed]