The Glaucoma-associated Olfactomedin Domain of Myocilin Is a Novel Calcium Binding Protein

Article

Shannon E. Hi

0022-2836/$ - see front m

The Glaucoma-Associated OlfactomedinDomain of Myocilin Forms PolymorphicFibrils That Are Constrained by PartialUnfolding and Peptide Sequence

ll, Rebecca K. Donegan an
d Raquel L. Lieberman
School of Chemistry and Biochemistry, Georgia Institute of Technology, 901 Atlantic Drive Northwest, Atlanta, GA 30332-0400, USA

Correspondence to Raquel L. Lieberman: [email protected]://dx.doi.org/10.1016/j.jmb.2013.12.002Edited by R. Wetzel

Abstract

The glaucoma-associated olfactomedin domain of myocilin (myoc-OLF) is a recent addition to the growing listof disease-associated amyloidogenic proteins. Inherited, disease-causing myocilin variants aggregateintracellularly instead of being secreted to the trabecular meshwork, which is a scenario toxic to trabecularmeshwork cells and leads to early onset of ocular hypertension, the major risk factor for glaucoma. Here wesystematically structurally and biophysically dissected myoc-OLF to better understand its amyloidogenesis.Under mildly destabilizing conditions, wild-type myoc-OLF adopts non-native structures that readily fibrillizewhen incubated at a temperature just below the transition for tertiary unfolding. With buffers at physiologicalpH, two main endpoint fibril morphologies are observed: (a) straight fibrils common to many amyloids and(b) unique micron-length, ~300 nm or larger diameter, species that lasso oligomers, which also exhibitclassical spectroscopic amyloid signatures. Three disease-causing variants investigated herein exhibitnon-native tertiary structures under physiological conditions, leading to a variety of growth rates and a fibrilmorphologies. In particular, the well-documented D380A variant, which lacks calcium, forms large circularfibrils. Two amyloid-forming peptide stretches have been identified, one for each of the main fibrilmorphologies observed. Our study places myoc-OLF within the larger landscape of the amylome andprovides insight into the diversity of myoc-OLF aggregation that plays a role in glaucoma pathogenesis.

© 2013 Elsevier Ltd. All rights reserved.

Introduction

Myocilin glaucoma is a recent addition to the familyof protein misfolding disorders. When it harbors oneof numerous non-synonymous mutations, myocilin,the gene product most closely linked to inheritedforms of primary open glaucoma [1], forms cytotoxicaggregates [2–5] within human trabecular meshwork(TM) cells located in the anterior segment of the eye.The toxic gain-of-function pathophysiology leadingto the early onset of ocular hypertension, the majorrisk factor for glaucoma that leads to retina degen-eration, is not well understood. On a cellular level,mutant myocilin interacts with endoplasmic reticulum(ER)-resident chaperones in a way that sabotagesits ability to be retrotranslocated and then cleared bythe proteasome [6]. Upon human TM cell death,debris is presumed to block outflow of fluid throughthe TM tissue, which in turn affects intraocularpressure. Elevated levels of wild-type myocilin are

atter © 2013 Elsevier Ltd. All rights reserve

also associated with glaucoma induced by steroidtreatment [7], but the pathogenic mechanism for thisglaucoma subtype is likewise unclear. Biophysicalcharacterization of the olfactomedin domain ofmyocilin (myoc-OLF) [8], the location of nearly allknown glaucoma-causing mutations [9,10], hasrevealed that such mutations lead to decreasedthermal stability that correlate with age of diseasediagnosis [11,12]. Moreover, aggregates of thiscalcium-binding [13] domain were recently found toexhibit hallmarks of amyloid fibrils in vitro, andsequestered mutant myocilin forms thioflavin T(ThT)-positive aggregates in cell culture [14].Viewed through the new lens of disease-relevant

amyloid proteins, which share a non-native cross-βcore structure [15], myocilin has intriguing features.First, mutant myocilin cytotoxicity originates inaggregation of mutant myocilin within the ER[3,16], whereas amyloid proteins are generallyfound to accumulate in the cytosol or extracellularly

d. J. Mol. Biol. (2014) 426, 921–935

http://dx.doi.org/

mailto:[email protected]

Fig. 1. Biophysical characterization of wild-type myoc-OLF as a function of pH, without NaCl. (a) Secondary structure ofmyoc-OLF as a function of pHmeasured by far-UV CD. (b) Tertiary structure of myoc-OLF as a function of pHmeasured bynear-UV CD. (c) Thermal unfolding of secondary structure monitored by CD at 214 nm. Fit is sigmoidal. (d) Thermalunfolding of tertiary structure monitored by CD at 291 nm. Fit is linear plus sigmoidal. (e) ANS fluorescence as a function ofpH. (f) ThT fluorescence as a function of pH at 36 °C monitored for 18 h. (g) Fibrillar endpoint morphology for samplesincubated at 36 °C in pH 3.4 buffer. (h) Disordered aggregates from samples incubated at pH 2.0 buffer at 36 °C, upondeposition (see the text). For (g) and (h), the scale bar in two left panels represents 300 nm and the scale bar in two rightpanels represents 50 nm.

922 Fibrils of the Myocilin Olfactomedin Domain

[17]. Second, although glaucoma is a neurodegener-ative disease of the retina leading to irreversible visionloss [18], thus far myocilin has not been directlyimplicated in neurotoxicity seen for numerous otherwell-studied amyloid proteins. Relatedly, myocilin isexpressed in tissues throughout the body and isconserved among higher eukaryotes where paralogsare expressed in neural tissues [19]; aggregation insuch tissues may be relevant but is currentlyunexplored. Third, beyond intracellular aggregationof mutant myocilin, extracellular, wild-type myocilin issubjected to the full complement of age-onset

misfolding triggers: repeated mechanical insults, pHimbalance, UV radiation, and reactive oxygen species[20,21]. Even though such conditions induce myoc-OLF to fibrillize in vitro [14] and glaucoma is highlyprevalent worldwide [22], the vastmajority of the agingpopulation does not have glaucoma, suggesting thatmyocilin can generally withstand these insults in theextracellular milieu. Fourth, recombinant ~30-kDamyoc-OLF is well-folded, thermally stable protein [8]and does not aggregate under physiological pH andtemperature except with agitation [14]. The corestructural domain, ~25 kDa, contains high levels of

Table 1. Summary of biophysical properties measured for myoc-OLF samples in this study

Secondary structureTm (°C)

Tertiary structureTm (°C)

ANS Fl ΔFThT/Foa Fibril

growth t1/2 (h)Fibril morphology

Wild-type myoc-OLFpH 7.2 51.5 ± 0.1 48.9 ± 0.6 0.9 ± 0.5 2.4 —

70.6b 2.0 StraightpH 5.8 52.7 ± 0.1 52.0 ± 0.4 1.1 ± 0.7 −0.4 —pH 4.6 48.7 ± 0.1 46.0 ± 0.3 10.6 ± 0.4 2.3 —pH 3.4 34.0 ± 0.6 29.0 ± 1.3 181.5 ± 2.3 79.8 0.4 ± 0.2 Long, straightpH 2 N/A N/A 298.3 ± 6.9 −1.7 —pH 7.2/NaCl 52.2 ± 0.6 50.9 ± 0.2 1.1 ± 1.1 −0.7

89.4b 27.7 ± 2.6 CircularpH 5.8/NaCl 54.0 ± 0.1 52.8 ± 0.3 1.4 ± 1.4 −0.5 —pH 4.6/NaCl 46.4 ± 0.1 43.1 ± 0.2 4.5 ± 2.3 −0.5 9.5 ± 3.9 Long, straightpH 3.4/NaCl N/A N/A 234.7 ± 3.9 21.5 —pH 2/NaCl N/A N/A 346.8 ± 24.0 16.0 —

myoc-OLF variantsI499F 45.0 ± 1.1 43.2 ± 0.5 3.4 ± 1.1 88.0 3.5 ± 0.2 Straight fibrils and

oligomersD380A 46.8 ± 0.1 44.4 ± 0.6 2.1 ± 0.4 102.2 11.9 ± 4.7 CircularA427T 48.9 ± 0.1 45.9 ± 0.2 1.3 ± 0.1 28.5 N25c Straight fibrils and

oligomersK398R 51.1 ± 0.4 49.5 ± 2.4 0.6 ± 0.2 −0.7 —

Data in boldface, fibrillar structures confirmed by AFM. N/A, not available because secondary structure is already disordered at 4 °C.a ThT fluorescence recorded after incubation for 18 h (no salt) or 50 h (200 mM NaCl) at 36 °C.b ThT fluorescence recorded after incubation for 18 h (no salt) or 90 h (200 mM NaCl) at 42 °C.c Steady-state maximum not reached by termination of experiment at 50 h.

923Fibrils of the Myocilin Olfactomedin Domain

β-sheet secondary structure [8] in a yet-unknown,three-dimensional architecture. The remaining~5 kDa is sensitive to proteolysis, and circulardichroism (CD) deconvolution corroborates that themyoc-OLF contains a significant contribution fromunstructured polypeptide [8]. Amyloid formation is wellknown for both folded and natively unfolded proteins,and the combination of the two in a single domainpresents an interesting platform to study nuances ofamyloidogenesis.To gain further insight into myocilin amyloidogen-

esis, we undertook a detailed molecular biophysicalstudy of myoc-OLF. Here we describe a spectrum ofpopulated states of wild-type myoc-OLF from nativeto acid-induced unfolded and identify key featuresand the conditions prone to fibrillization or non-specific aggregation. Of the states that readily formfibrils, different experimental conditions such as pH,salt, temperature, and details of experimentalmaterials and equipment lead to different fibrilmorphologies. Two main structures have beenidentified by atomic force microscopy (AFM), typicallong straight fibrils and a highly unusual, micro-n-length circular species that encase oligomers.Both fibril forms are ThT positive, are Congo Redsensitive, and exhibit vibrational spectroscopicsignatures of amyloid. The three disease-causingmyoc-OLF variants investigated here form analo-gous fibrils under physiological buffers and temper-atures; the calcium-depleted D380A variant formsthe circular morphology. Two peptide stretches

identified by bioinformatics fibrillize into thesedisparate structures, suggesting that they containthe core sequences responsible for these fibril typesin myoc-OLF. Besides expanding our knowledge onthe amylome superfamily, our findings raise somenew issues regarding myoci l in glaucomapathogenesis.

Results

Overview

We systematically perturbed the myoc-OLFstructure to identify features that lead to facileamyloid formation. Structural and stability changesof wild-type myoc-OLF at five pH values above andbelow the calculated pI (~5) were monitored by far-and near-UV CD spectra and thermal melts. Effectsdue to charge screening were accounted for bycomparing results with and without 200 mM NaCl.The corresponding extent of exposed hydrophobicitywas tested using anilinonaphthalene-1-sulfonate(ANS) fluorescence at room temperature. To relatethe identified partial unfolded states with amyloidpropensity, we monitored kinetics of fibril formationusing the fluorescent amyloid indicator dye ThT[23]. Experiments were carried out over the courseof 18–90 h in a cuvette or plate reader held at 36 °Cto approximate the physiological temperature of theeye [24] or at a slightly elevated temperature of

Fig. 2. Biophysical characterization of wild-type myoc-OLF as a function of pH, in the presence of 200 mM NaCl.(a) Secondary structure of myoc-OLF as a function of pH measured by far-UV CD. (b) Tertiary structure of myoc-OLFas a function of pHmeasured by near-UVCD. (c) Thermal unfolding of secondary structuremonitored byCDat 214 nm. Fit issigmoidal. (d) Thermal unfolding of tertiary structure monitored by CD at 292 nm. Fit is linear plus sigmoidal. (e) ANSfluorescence as a function of pH. (f) ThT fluorescence as a function of pH at 36 °C monitored for 50 h. (g) Fibrillar endpointmorphology for samples incubated at 36 °C in pH 4.6 buffer. (h) Deposits of curvilinear fibrils fromsamples incubated at 36 °Cin pH 3.4 buffer. (i) Disordered aggregates for samples incubated at 36 °C in pH 2.0 buffer. For (g)–(i), the scale bar in two leftpanels represents 300 nm and the scale bar in two right panels represents 50 nm.


42 °C (see below). Endpoint morphologies werethen visualized by AFM. Neutral pH wild-type fibrilswere further characterized by Fourier transforminfrared spectroscopy (FTIR) and Congo Red

absorbance. In addition to wild-type myoc-OLF,experiments were conducted on four representativevariants and selected peptides at physiological pHand temperature.

image of Fig.�2


pH screen of wild-type myoc-OLF fibrillization inbuffers lacking NaCl

The secondary structure of myoc-OLF is relativelyunchanged over the pH range 7.2–3.4, but notablechanges in tertiary structure are seen. Across this pHrange, the far-UV CD spectrum exhibits the same,previously reported [8] minimum around 215–217 nmtypical of β-sheet structure and a shoulder at 230 nm(Fig. 1a). The more pronounced 230-nm trough in thepH 4.6 and pH 3.4 spectra is consistent with changesto the environment of aromatic residues [25], asdetailed in the near-UV spectrum where the shoulderbetween 280 and 250 nm [26,27] appears to broaden(Fig. 1b). At pH 2,myoc-OLF is reminiscent of an acid-induced unfolded state as evidenced by the alteredsecondary structure in the far-UV CD spectrum(Fig. 1a) and loss of tertiary structure indicated bythe flat near-UV CD signal [28,29] (Fig. 1b). CDthermal melts reveal that, whereas secondary struc-ture unfolding appears to follow a sigmoidal transition(Fig. 1c), tertiary thermal transitions exhibit a shallowpre-transition phase followed by a sharp main unfold-ing event (Fig. 1d). Thermal melts following changesin intrinsic tryptophan fluorescence confirm thispre-transition slope, with a 0.4-nm shift in maximumfluorescence to lower energies every 10 °C (Supple-mental Fig. S1). Myoc-OLF is least stable in acidicbuffers where the protein is highly positively charged,and thermal stability decreases as the protein isshifted to buffers with pH values below its predictedpI (Table 1). ANS fluorescence is highest formyoc-OLF at pH 2 and pH 3.4, consistent withhigh levels of exposed hydrophobic residues(Fig. 1e). ANS measurements reveal a 17-foldhigher fluorescence (Table 1) for myoc-OLF desta-bilized at pH 3.4 compared to the more compactstate at pH 4.6 (Fig. 1e).Of the conditions tested, only the distinct partially

unfolded state accessible at pH 3.4 is primed forfibril formation at 36 °C (Fig. 1f). None of the otherconditions led to fibril growth at this temperature,including the acid-induced unfolded state at pH 2and the natively folded state at pH 7.2 (Fig. 1f).Corresponding AFM images of fibrils grown inpH 3.4 buffer at 36 °C reveal characteristic straightamyloid morphology (average contour length,271 ± 164 nm; height, 0.9 ± 0.4 nm; actual width,8 ± 2 nm), as well as less ordered, laterally assem-bling oligomers (average contour length, 126 ± 71 nm;height, 2.0 ± 0.5 nm; actual width, 10 ± 2 nm)(Fig. 1g). AFM images confirm that no fibrils grewfrom experiments conducted at pH 7.2, pH 5.8, orpH 4.6 at 36 °C (Supplemental Fig. S2a–c). Finally,although samples deposited from conditions at pH 2appear as disordered aggregate (Fig. 1h), webelieve that this is an artifact of the deposition ofunstructured monomers on the mica surface be-cause the average particle height (4.2 ± 0.3 nm and

1.3 ± 0.4 nm) is comparable to monomeric myo-c-OLF prior to aggregation (Supplemental Fig. S3)and no increase in scattering at 600 nm during theincubation period was seen (Supplemental Fig. S4).Shorter fibril heights of ~1 nm, such as those seenin AFM images at pH 3.4 (Fig. 1g), are consistentwith dehydrated myoc-OLF monomer deposited onmica for AFM (Supplemental Fig. S3b and c).

pH screen of wild-type myoc-OLF fibrillization inbuffers supplemented with 200 mM NaCl

Similar to results obtainedwith buffers lackingNaCl,the secondary structure of wild-type myoc-OLF isunchanged formyoc-OLF in buffers at pH 7.2, pH 5.8,and pH 4.6 with 200 mM NaCl (Fig. 2a). For pH 4.6,there is a slight increase in the feature at 230 nm,again suggesting a change in the aromatic environ-ment (Fig. 2a). This interpretation is supported by asignificantly broadened shoulder in the near-UV CDspectrum between 280 and 250 nm [26,27] (Fig. 2aand b). Unlike the parallel case without salt, theaddition of 200 mMNaCl at pH 3.4 leads to extensivedisorder in tertiary structure and altered secondarystructure (Fig. 2a and b). For pH 2, the presence of200 mM NaCl does not change the extent of ordercompared to samples incubated without salt (Fig. 2aand b). Secondary structure melts are best modeledas sigmoidal with a melting temperature severaldegrees Celsius higher than tertiary unfolding,which, as without salt (Fig. 1c and d), follows ashallow pre-transition step before the main unfoldingevent (Fig. 2c and d). In comparison to that in theabsence of NaCl, the pre-transition feature in thepresence of 200 mM NaCl is less pronounced, mostlikely due to increased electrostatic screening ofcharged residues [30] (Figs. 1d and 2d). CD resultsare corroborated by ANS fluorescence, which ishighest for pH 2 and pH 3.4 and is low for all otherpH values, including pH 4.6 (Fig. 2e and Table 1).Fibril assays (Fig. 2f) combined with AFM imaging

(Fig. 2g) reveal that when salt is present, only thepartially unfolded state accessed at pH 4.6 is primedfor typical fibril growth. Over the 2-day incubation,there is an exponential increase in ThT fluorescenceseen alongwith primarily long, straight fibrils by AFM(average contour length, 2643 ± 600 nm; height,1.1 ± 0.5 nm; actual width, 12 ± 2 nm) (Fig. 2g).Notably, myoc-OLF samples at pH 3.4 and pH 2showed enhanced ThT fluorescence at the start ofthe incubation, but no significant change is observedover time. Analysis by size-exclusion chromatographyconfirms that myoc-OLF incubated at pH 3.4/200 mMNaCl is aggregated at the start of the incubation whilemyoc-OLFatpH 4.6/200 mMNaCl remainsmonomeric(Supplemental Fig. S5). Images from incubation atpH 3.4 reveal some fibril-like aggregates (averagecontour length, 127 ± 61 nm; height, 0.8 ± 0.3 nm;actual width, 9 ± 2 nm), but many are clustered

Fig. 3. Fibrillization of wild-type myoc-OLF incubated at 42 °C in neutral pH buffer. (a) ThT fibril growth curves includingscattering intensity kinetics for incubation at 42 °C without NaCl and (b) with 200 mM NaCl. (c) Extended fibril morphologyobserved for samples incubated in the absence of NaCl. (d) Loop morphology observed for samples incubated in thepresence of 200 mM NaCl. The scale bar for both conditions (c and d) is 300 nm. (e) FTIR spectral analysis for monomersand aggregates without NaCl and (f) with 200 mM NaCl. Dashed lines, Gaussian fits. Bold line, sum of Gaussian fits.


together into large disordered structures (Fig. 2h). Theunfolded state at pH 2 exhibits a wide distribution ofheights ranging from 10.8 to 42.2 nm suggestive ofextensive aggregation in the sample prior to deposition(Fig. 2i). The two remaining myoc-OLF samples, atpH 7.2 and pH 5.8, do not form fibrils, even after the2-day incubation at 36 °C (Supplemental Fig. S2d ande).

Fibrillization of wild-type myoc-OLF at neutralpH and elevated temperatures

Although myoc-OLF does not readily form fibrils atneutral pH when incubated at 36 °C (Figs. 1f and 2f)

or at 37 °C without agitation [14], myoc-OLF canform fibrils when incubated at 42 °C, the temperatureat the onset of the main tertiary unfolding transition(Figs. 1d and 2d; Supplemental Fig. 6b). In theabsence of NaCl, the time to half-maximal ThTfluorescence (t1/2) is 2.0 h, whereas fibrillization isconsiderably slower with the addition of 200 mMNaCl, with a t1/2 of 27.7 h (Fig. 3a and b; Table 1).The increased ThT aggregation rate for wild-typemyoc-OLF in the absence of NaCl is likely because ofits lower stability (Table 1) due to charge repulsion andnot becauseof an intrinsic difference in the structure at42 °C prior to aggregation (Supplemental Fig. 6a). Inaddition, light-scattering kinetics in the absence of

image of Fig.�3

Table 2. FTIR analysis of β-sheet content for myoc-OLF wild-type monomer and aggregates

Amide I′ maximum (cm−1) Native β-sheet (%) Amyloid β-sheet (%) Turns (%) Disorder (%)

Monomer (7.2) 1635 75 — 25 —Monomer (7.2/NaCl) 1635 74 — 26 —Aggregate (7.2) 1621 — 68 18 14Aggregate (7.2/NaCl) 1621 — 75 18 7

Native β-sheet bands were assigned from 1630 to 1641 cm−1 while amyloid β-sheet bands were assigned from 1611 to 1630 cm−1 asdescribed in Ref. [31].


NaCl show a typical sigmoidal growth curve foramyloid fibrils, with a lag phase of ~5 h (Fig. 3a). Inthe presence of salt, the lag phase is reduced to30 min or less (Fig. 3b) with an increase in the ratio ofscattering intensity to ThT fluorescence, suggestingan increase in non-specific aggregation (Fig. 1b). Forboth samples, endpoint fibrillar structures are seen byAFM upon the termination of the fibrillization experi-ment after 90 h, but unexpectedly, the presence orabsence of salt in the buffer leads to differentmorphologies. In the absence of NaCl, a mixture offibrils and oligomers are present (fibril averagecontour length, 334 ± 289 nm; height, 1.0 ± 0.3 nm)(Fig. 3c), while in the presence of 200 mM NaCl,the predominant species appears as oligomers(3.9 ± 1.9 nm in height) enclosed by closed, ornearly closed, loop fibril structure of 10.9 ± 2.5 nmin height and microns in length (an averagecontour length of 2440 ± 1240 nm and an averageend-to-end distance of 48 ± 35 nm) (Fig. 3d andSupplemental Fig. S7). Both morphologies exhibit theexpected amyloid-associated shift in Congo Redabsorbance (Supplemental Fig. S8). FTIR spectraalso display key shifts in the amide I region from theprototypical native β-sheet signature in themonomericsamples at 1635 cm−1 compared to the correspond-ing amyloid feature at 1621 cm−1 observed in theaggregated samples (Fig. 3e and f). Deconvolution ofthe smoothed spectra in this region reveals that, forboth straight and curved morphologies, ~70% of thesecondary structure in the aggregates is amyloid(Table 2).

Non-native structure, fibril growth, and fibrilmorphologies of myoc-OLF variants underphysiological conditions

Next, four representative myoc-OLF variants [9]were evaluated for their structure, stability, andamyloid fibrillization propensity under physiologicalconditions (pH 7.2/200 mM NaCl, 36 °C): (a) K398R,a wild-type-like single-nucleotide polymorphism vari-ant [12]; (b) A427T, a disease-causing variant withnear-wild-type stability [12]; (c) D380A, a disease-causing variant lacking a calcium ion cofactor [13];and (d) I499F, a disease-causing variant withsignificantly reduced thermal stability compared to

wild type [12]. Purity and monomeric state of eachmyoc-OLF variant were confirmed by size-exclusionchromatography [12] and SDS-PAGE (SupplementalFig. S9). For all four variants, whereas secondarystructure CD spectra are nearly superimposable withwild type (Fig. 4a), the three disease-causing variantsA427T, D380A, and I499F exhibit significant non-native tertiary features (Fig. 4b). The tertiarystructure of K398R is indistinguishable from wildtype (Fig. 4b), but both A427T and D380A havebroader negative features in the phenylalaninerange (270–250 nm) [26] (Fig. 4b), correlating withtheir reduced thermal stability (Table 1). The I499Fvariant is likewise non-native but positive in thisregion of the near-UV spectrum (Fig. 4b), whichmay be due to the additional phenylalanineresidue. Like wild-type myoc-OLF, unfolding oftertiary structure for all four variants involves apre-transition followed by a major unfolding event(Fig. 4d); secondary structures melt in an apparentsigmoidal transition 2–3 °C higher (Fig. 4c andTable 1). Melting temperatures are consistent withpreviously reported values determined by othermethods [12,13]. Comparison of ANS fluorescencelevels reveals that the three disease variants areless compact proteins compared to wild-typemyoc-OLF or K398R to an extent comparablewith wild type at pH 4.6/200 mM NaCl but areless unfolded than wild type at pH 2 (Table 1;Figs. 2e and 4e).Kinetics of amyloid formation were evaluated next

for each variant at 36 °C. For K398R, low constantlevels of ThT fluorescence indicate no fibrillization, likewild-type myoc-OLF under these conditions. Bycontrast, all three disease-causing variants formThT-positive structures (Fig. 4f). The less stablevariants fibrillize faster, as measured by t1/2 (Fig. 4fand Table 1). AFM imaging confirms that A427T,D380A, and I499F aggregate into diverse, amyloid-likemorphologies (Fig. 4g–i), while the K398R does not(Supplemental Fig. S2f). For A427T and I499F, asignificant population of the aggregates appears asoligomers, as well as a few fibrils with heights andcontour lengths of 3.3 ± 0.4 nm and 166 ± 101 nm(A427T) and 3.0 ± 0.4 nm and 282 ± 143 nm (I499F),respectively. Fibrils of D380A resemble wild-typemyoc-OLF incubated at pH 7.2 with salt at 42 °C

Fig. 4. Biophysical characterization of myoc-OLF variants K398R (SNP), D380A, A427T, and I499F (disease causing)in physiological buffers. (a) Secondary structure measured by far-UV CD. (b) Tertiary structure measured by near-UV CD.(c) Thermal unfolding of secondary structure monitored by CD at 214 nm. Fit is sigmoidal. (d) Thermal unfolding of tertiarystructure monitored by CD at 292 nm. Fit is linear plus sigmoidal. For (a)–(d), wild-type myoc-OLF at neutral pH with200 mM NaCl is overlayed for comparison. (e) ANS fluorescence as a function of pH, with overlay of wild-type myoc-OLFat pH 7.2/200 mM NaCl and 4.6/200 mM NaCl. (f) ThT fluorescence at 36 °C monitored for 50 h. (g) Endpointmorphologies seen for myoc-OLF(A427T) are fibrils and oligomers. (h) Deposits of myoc-OLF(D380A) appear ascurvilinear and circular fibrils enclosing smaller globular aggregates. (i) Morphologies seen for myoc-OLF(I499F) are fibrilsand oligomers, similar to myoc-OLF(A427T). For (g)–(i), the scale bar in two left panels represents 300 nm and the scalebar in two right panels represents 50 nm.


(Fig. 3d), appearing as closed or nearly closed fibrilloops surrounding oligomers (Fig. 4h). The oligomershave a height of 1.2 ± 0.3 nm and a surrounding fibrilborder that is 1.4 ± 0.4 nm tall with an average contourlength of 2119 ± 887 nm.

Identification of amyloidogenic peptides withinthe myoc-OLF sequence

To assess whether different amino acid se-quences within myoc-OLF could be responsible for

image of Fig.�4


forming the two major mature aggregate morphol-ogies observed, namely, long straight/curvilinearfibrils (Figs. 1g, 2g, 3c, and 4g and i) or lassoedoligomers (Figs. 3d and 4h), we evaluated theamino acid sequence myoc-OLF using fouramyloid prediction algorithms, Waltz [32], PASTA[33], AmylPred [34], and TANGO [35]. The threehighest scoring sequences, G326AVVYSGSLYFQ(P1), G387LWVIYSTDEAKGAIVLSK (P2), andV426ANAFIICGTLYTVSSY (P3) (Supplemental Fig.S10), were identified with consensus priority foramyloid of P3 N P1 N P2 and were synthesized forfurther evaluation. Indeed, incubation of P1andP3, butnot P2 for 20 h at 36 °C, reveals a high level of ThTfluorescence (Fig. 5a). Additional incubation of P2 for72 h still did not lead to a change in ThT fluorescence(data not shown).While the growth of P1 is exponentialover the course of the experiment, rates could not beslowed to observe a clear lag phase. P3 fibrillizationis nearly instantaneous upon dilution of the DMSO-dissolved peptide into buffer, and could not beslowed by altering experimental conditions such asfurther dilution, temperature, or buffer. By AFM, P1appears as straight amyloid fibrils, with an averageheight and contour length of 0.9 ± 0.2 nm and700 ± 537 nm (Fig. 5b) akin to myoc-OLF(Figs. 1g, 2g, and 3c); P2 is not aggregated(Fig. 5c), and P3 aggregates into closed amyloidfibrils with an average height and contour length of1.4 ± 0.6 nm and 1566 ± 298 nm (Fig. 5d), similar towild-type myoc-OLF in pH 7.2/NaCl and 42 °C(Fig. 3d) and D380A (Fig. 4h).

Discussion

Amyloid formation is a kinetically driven, self-assembly process that results in common characteristicfilaments [36] among disparate proteins [37]. Oftenassociated with diseases, including Alzheimer,Parkinson, andHuntington, where they are neurotoxic,amyloid can also play a functional role [17], mayrepresent an ancient protein fold [38], and can beformed by model proteins [39]. Amyloids arebelieved to share a core interdigitated, so-calledcross-β steric zipper [40,41] that forms via anaccessible partially folded state and is controlledby colloidal stability and kinetics [37]. Bioinformaticsapproaches can evaluate sequences with highamyloid propensity from chemical aspects ofamino acid sequence [42]. However, observedsupramolecular morphologies vary widely and arecritically dependent on the precise experimentalconditions or amino acid substitutions that perturbprotein structures to enable access to one or moreamyloidogenic peptide stretches [23,30,43–45].To address the molecular details of amyloid forma-

tion by myoc-OLF, a newly identified amyloid protein[14], and enable comparison with other highly studied

amyloids, we identified and characterized a spectrumof populated conformations. The native-state CDsignature involves previously identified secondarystructural features at ~215 nm and a 230 nm shoulder[8]; here we add the additional characteristic of nativetertiary structure, namely tight double troughs at 282and 291 nm. Native-state CD spectra are superim-posable regardless of the presence of salt. At the otherextreme is the acid-induced unfolded state that lackstertiary structure,maintains some secondary structure,but has a high level of ANS fluorescence (Fig. 1). In theabsence of salt at pH 2, the high positive charge (26Arg, Lys residues) leads to charge repulsion thatenables an unfolded state to exist that does notaggregate or form fibrils. The addition of salt at pH 2(Fig. 2) introduces charge shielding and attractiveintermolecular forces commonly described by thesecond virial coefficient B22 that lead to immediatebulk, disordered aggregation [30,46,47].At intermediate pH values, we have identified a

spectrum of partially unfolded myoc-OLF states, andit is these non-native, destabilized states that areprone to fibril formation under distinct experimentalconditions. In the presence of NaCl, we character-ized one partially folded state with altered tertiarystructure at pH 4.6; at other pH values, myoc-OLFwas either a largely, aggregation-prone unfoldedstate or indistinguishable from its native conforma-tion (Fig. 2). In the absence of NaCl, we identifiedseveral non-native states from pH 3.4 to pH 5.8(Fig. 1), likely due to decreased charge screening[47]. All of these states are less thermally stable thannative but of variable levels of exposed hydrophobicenvironments. Fibrillization can proceed from manyof these states under our non-agitating assayconditions when incubated at a temperature nearthe onset of tertiary unfolding [46].The importance of the non-native and destabilized

conformational state requirements for fibril formationunder conditions of favorable intermolecular interac-tions is underscored by our results for wild-typemyoc-OLF and variants under physiological condi-tions. In the case of pH 7.2/NaCl, wild-type myoc-OLFwill fibrillize without agitation at 42 °C, the onset oftertiary unfolding of myoc-OLF at this pH value. Wepreviously reported [14] that incubation at 37 °C for atleast 2 weeks does not lead to fibril formation [14].Rather, to grow fibrils at 37 °C, agitation is required[14], which likely introduces a combination of mechan-ical local unfolding, effects due to the liquid/airinterface, and concentration fluctuations that createnuclei for amyloid formation [39,48,49]. We alsodemonstrated that other destabilizing reagents suchas acid spike or peroxide, which mimic agingconditions, presumably act similarly to acceleratefibrillization of myoc-OLF under conditions that itwould otherwise not form fibrils [14]. The threedisease-causing myoc-OLF variants, A427T, D380A,and I499F, investigated in this study (Table 1) are

Fig. 5. Fibrillization of peptides P1–P3 incubated at 36 °C in physiological buffer. (a) ThT fibril growth kinetics. (b) Straightfibrilmorphology observed for P1. (c) Noaggregates observed for P2. (d)Closed loopmorphologyobserved for P3. Scale barsfor (b)–(d) are 300 nm.


destabilized to variable extents under physiologicalbuffer and temperatures. They exhibit a non-nativetertiary structure such that fibrils can form at 36 °C, atemperature consistent with that found in the eye, andwithin the window where fibrils would be expected toform along the unfolding pathway based on our pHstudy. We infer that fibrillization at 36 °C is a commonfeature of the many other disease-causing variantswhose measured Tm values [12] are bracketed by thebroad spectrum represented by A427T, D380A, andI499F.A variety of fibril morphologies were observed in

this study under different experimental conditions,including an unprecedented endpoint structure of alarge fibrillar closed ring that appears to encapsulateoligomers. Myoc-OLF aggregate morphology ishighly dependent on pH, salt, and temperature ofincubation. In several conditions, myoc-OLF formsstraight fibrils expected for amyloids, with dimen-sions common to other amyloid fibrils [50]. Asillustrated for fibrils grown from wild-type myoc-OLFat pH 7.2 without salt after incubation at 42 °C,amyloid signatures in FTIR and the expected shift ofCongo Red absorbance are present. For the micronlength fibril loops with enclosed oligomers grown inthe presence of salt at neutral pH at 42 °C [30,43],the amyloid character was confirmed by a CongoRed absorbance shift and signature resonancesin FTIR, but additional studies will be necessaryto clarify its three-dimensional structure. Annularaggregates have been reported for a few otheramyloid proteins including equine lysozyme [51] andα-synuclein [52], but these are less than 100 nm incircumference and are short-lived intermediates onthe pathway to fibril formation. To the best of ourknowledge, there is just one other protein reported toform fibrils with any resemblance to these structures,

the non-disease-associated N-terminal domain ofEscherichia coli HypF (HypF-N) [53]. HypF-N fibrilsare ThT-positive crescents of similar size tomyoc-OLF, but there are numerous differences.First, the predominant structure is crescent-shapedand devoid of material in the center; closed rings,when observed, are transient. Second, HypF-Ncrescents grow from 30% TFE at pH 5.5 over thecourse of 3 days at room temperature, unlike ourmild and more biological conditions. Finally,HypF-N crescents or rings disappear and formtwisted long ribbons by 5 days, whereas closed-ringmyoc-OLF fibrils do not morph into such ribbon-likefibrils. Unexpectedly, in addition to wild-type myoc-OLF, the D380A disease variant, which is depletedof its calcium ion [13], also forms these ring-likestructures. Emerging evidence from other amyloidssuch as equine lysozyme [51] and gelsolin [54]support the hypothesis that metal binding, andcalcium in particular, affects fibril formation andmorphology by changing local protein structure and/or electrostatics that affect intermolecular interac-tions; both could be in play in the case of D380A orother variants that affect calcium binding.Individual prediction programs vary widely on how

many amyloidogenic sequences myoc-OLF harborsin total, but inspection of the program outputs side byside led to the selection of our three top candidatepeptides, P1–P3, which have predicted pI values of5.5–6.0 and, thus, like myoc-OLF, would be nega-tively charged at neutral pH. P1 forms long straightfibrils of typical amyloid morphologies similar tomyoc-OLF at pH 7.2 without salt, exhibiting exponen-tial growth upon dilution into physiological buffer at36 °C. P3 aggregated upon dilution, forming largerings similar tomyoc-OLF at pH 7.2/NaCl and D380A.These data support the hypothesis that the two main

image of Fig.�5


morphologies seen experimentally derive from thesepeptide stretches, although we cannot exclude otheramyloidogenic peptide stretches within myoc-OLF.While it is surprising that P2 did not form fibrils giventhe high level of confidence in its prediction,particularly for the N-terminal sequence GLWVIY,perhaps this is because P2 has the least overallpredicted secondary structure. The ring structuresseenwithP3 suggest that the fibril core ismore flexibleor kinked than in P1. Perhaps this is a consequence ofthe presence of an internal Cys; inmyoc-OLF, there isa disulfide bond involving this Cys, which mayinfluence the unfolding pathway under certain exper-imental unfolding scenarios. Computational predictionusing PASTA [33] also indicates that P3 may formantiparallel β-strands, compared to P1, for whichparallel strands are more likely. Further studies toelucidate the closed-ring morphology, including thestructures of the encased oligomers apparent in bothmyoc-OLF and D380A, are warranted.In sum, myoc-OLF is a newly identified, disease-

related protein that can form bona fide amyloid fibrilswith a variety of morphologies initiated by access of apartially folded state via particular experimentalparameters. The finding that the three disease variantsexamined here formdifferent fibril morphologies raisessome questions about the molecular recognition ofmutant myocilin by ER-resident chaperones such asGrp94 [6]. Specifically, it is of interest whether suchrecognition is sequence dependent, and if so, if allknown disease mutants expose this sequence.Alternatively, chaperone recognition could be depen-dent on fibril morphology, requiring different chaper-ones depending on the variant. Regarding wild-typemyocilin, it is perplexing that a broad role foraggregation in glaucoma is currently only specula-tive given that fibrillogenesis is relatively facile invitro and given the fact that the physiologicalenvironment of the TM has abundant knownfacilitators of fibril formation such as glycosami-noglycans [55–57]. We would predict that mor-phologies we reported here could damage TMtissue and in turn influence the balance of fluidflow and intraocular pressure, akin to oligomers ofAβ, which form membrane-spanning channels thatrelease calcium ions [58]. If instead protectivemechanisms are at play in the TM that suppressextracellular myocilin aggregation, such knowledgecould be of high value to other amyloid systems.

Materials and Methods

Protein expression and purification of myoc-OLF

Myoc-OLF and variants were expressed as MBP-OLFfusion proteins and isolated after cleavage by Factor Xa,as previously described [8,11–13], or by cleavage with the

tobacco etch virus (TEV) protease. For cleavage with theTEV protease, the original pMAL-c4x construct with theFactor Xa cleavage site (SSIEGR) betweenMBP andOLFwasmutated to a TEV protease cleavage site (ENLFYQS)using the site-directed mutagenesis (QuikChange; Strata-gene) kit. Primers are listed in Supplemental Table S1. Allplasmids were verified by DNA sequencing (Operon).Expression and purification of MBP-OLF(TEV) proceededas previously described for MBP-OLF(Factor Xa). Cleav-age of MBP-OLF(TEV) was accomplished by overnight(~18 h) incubation with TEV protease at room tempera-ture in 10 mM Na2HPO4/KH2PO4 and 200 mM NaCl(pH 7.2) buffer. TEV protease was produced in-houseusing the pRK793 plasmid (Addgene) as describedpreviously [59]. After cleavage, the TEV protease(6×His-tag) was captured by nickel affinity using a 1-mLHis Trap (GE Healthcare), followed by removal of MBPwith amylose affinity chromatography and isolation ofmonomeric OLF by gel filtration using the Superdex 75(GE Healthcare).

Buffers for pH study

Myoc-OLF was examined in five buffers, with or without200 mM NaCl: 10 mM Na2HPO4/KH2PO4 (pH 7.2),10 mM Na2HPO4/KH2PO4 (pH 5.8), 10 mM sodiumacetate (pH 4.6), 10 mM citrate (pH 3.4), and 10 mMphosphate (KH2PO4/phosphoric acid; pH 2). Immediatelyprior to any experiment, purified myoc-OLF was concen-trated at 4 °C and re-diluted three times into theappropriate buffer using an Amicon Ultra-15 Centrifugationdevice. The final protein concentration was determinedafter buffer exchange using the predicted molar extinctioncoefficient at 280 nm of 68,425 M−1 cm−1. Myoc-OLFvariants were purified using 10 mM Na2HPO4/KH2PO4and 200 mM NaCl (pH 7.2) directly and used withoutfurther buffer exchange.

Circular dichroism

CD studies were conducted using a Jasco J-810 orJ-815 spectropolarimeter equipped with a Neslab RTE 111circulating water bath and a Jasco PTC-4245/15 temper-ature control system. Far-UV CD spectra were acquired ina 0.1-cm cuvette from 300 nm to 200 nm, at 4 °C, with ascan rate of 500 nm min−1 and a data pitch of 1 nm.Protein concentration was 10–12 μM, and for each buffer,the reported spectra are an average of duplicate mea-surements of 10 scans that were background-subtracted.Far-UV CD melts were performed by increasing thetemperature at 2 °C min−1 from 4 °C to 70 °C and adelay time of 60 s before each acquisition. The spectrumat each temperature is an average of 10 scans usingabove parameters. Averaged spectra were converted tomean residue ellipticity Θ = (Mres × Θobs)/(10 × d × c),where Mres = 112.9 is the mean residue mass calculatedfrom the protein sequence, Θobs is the observed ellipticity(°) at wavelength λ, d is the pathlength (cm), and c is theprotein concentration (g mL−1). The Tm was determinedusing mean residue ellipticity values recorded at 214 nmvia Boltzmann Sigmoid analysis in Igor Pro.Near-UV CD spectra were acquired using a 0.1-cm

cuvette at 4 °C, from 320 nm to 250 nm with a scan rate of


50 nm min−1 and a data pitch of 1 nm. Protein concen-tration was 40–50 μM, and the reported spectra for eachbuffer are an average of duplicate measurements from 10scans that were background-subtracted. Near-UV CDmelts, performed in duplicate, were conducted by increas-ing the temperature by 2 °C min−1 from 4 °C to 70 °C, witha delay time of 60 s before each acquisition. The spectrumat each temperature is an average of 10 scans usingaforementioned scan parameters. The near-UV spectrawere baseline-subtracted from an average spectrum valuecalculated from spectra acquired during the melt at eachtemperature, to account for aggregation observed due tohigher protein concentrations (range 320–310 nm). TheTm was determined using mean residue ellipticity values at291 nm or 292 nm and fit using a combination of a linearand sigmoidal functions in Igor Pro [60].

ANS fluorescence

ANS fluorescence was measured on a ShimadzuRF-530/PC spectrofluorophotometer at room tempera-ture, with an excitation wavelength of 380 nm (slit width,3 nm) and an emission range 400–600 nm (slit width,1.5 nm). Stock solutions of 10 mM ANS in deionizedwater were diluted to 600 μM working stock solutions inappropriate buffer. Each sample, composed of 10 μMmyoc-OLF and 100 μM ANS, was prepared 15 min priorto data acquisition. Five accumulations were averagedand background-subtracted for each measurement.Reported spectra are an average of two independentexperiments.

Fibrillization assay

Amyloid aggregation was monitored by ThT fluores-cence. Myoc-OLF and variants at 30 μM were incubatedover time in a variety of buffer conditions. A stock solutionof 1 mg mL−1 ThT in deionized water was diluted to200 μM working solutions in appropriate buffer anddispensed for a final concentration of 10 μM for aggrega-tion experiments. For experiments without NaCl, data wereacquired on a Shimadzu RF-530/PC spectrofluorophot-ometer using a 200-μL low-head space cell (Starna26.50LHS-Q-10/Z15). ThT fluorescence was recordedevery 10–30 min using an excitation wavelength of440 nm (slit width, 3 nm) and an emission range 450–600 nm (slit width, 3 nm). The change in the maximumThT intensity at 485 nm plotted versus time. Experimentsin the presence of NaCl were conducted using a BiotekSynergy microplate reader equipped with a 440-nmexcitation filter and a 485-nm emission filter. myoc-OLFsamples (150 μL) were prepared in 1.5-mL centrifugetubes, transferred to a 96-well microplate (Grenier), andthen sealed with clear MicroAmp PCR film (AppliedBiosystems). Measurements were recorded every 10 minand background-subtracted with a corresponding buffer-only solution supplemented with 10 μM ThT. At thetermination of the assay (18–90 h), samples wereremoved from the microplate and stored in microcentrifugetubes until imaging by AFM (see below). All presentedfibrillization assay data are an average of at least twoindependent experiments rescaled asΔF/Fo, where Fo is theinitial fluorescence. Aggregation was also monitored by

light scattering using a Zetasizer Nano S (MalvernInstruments, Worcestershire, UK) with a He-Ne laser(λ = 633 nm) or by monitoring absorbance (λ = 620 nm)using a Biotek Synergy microplate reader. All presentedlight-scattering data are rescaled as ΔI/Io, where Io is theinitial scattering intensity or initial absorbance.

Peptide synthesis and fibril formation

Peptides were synthesized by Celtek Peptides(Nashville, TN). Peptides G326AVVYSGSLYFQ (P1)and G387LWVIYSTDEAKGAIVLSK (P2) are N95%pure, and peptide V426ANAFIICGTLYTVSSY (P3) wasdesalted. Peptides were prepared as stock solutions inDMSO (5 mg mL−1) and stored at room temperature.For reported fibrillization assays, peptides were dilutedto 500 μM in 10 mM Na2HPO4/KH2PO4, 200 mM NaCl(pH 7.2), and 10 μM ThT. Aggregation was monitoredas mentioned above for experiments with myoc-OLF atpH 7.2 without salt.

Atomic force microscopy

A few hours after the termination of the aggregationassay, 40 μL was removed from the bottom of the 1.5-mLmicrocentrifuge tube and left to be adsorbed onto freshlycleaved mica for 30 min, rinsed for 3 s with deionizedwater, and left to dry overnight in a Petri dish. Formyoc-OLF at pH 3.4 and pH 2 in the absence of NaCl,pH 7.2 in the absence of NaCl at 42 °C,myoc-OLF(D380A),and P3, samples were diluted 100-fold prior to deposition.Dry samples were imaged in air with a MFP-3D atomicforce microscope (Asylum Research) using PPP-FMR(NanoAndMore) silicon tips with nominal tip radii less than7 nm. The cantilever was driven at 60–70 kHz inalternating current mode and a scan rate of 0.5 Hz with512 × 512 pixel resolution. Raw image data were correctedfor image bow and slope using the software provided byAsylum Research. Dimensional analysis for average heightand width was performed using the particle analysissoftware provided by Asylum Research, ignoring particleswith anarea less than25 nm2 andaheight less than0.5 nm.The actual width was deconvoluted from the averageapparent width using Wact = Wapp − 2[H(2Rt − H)]1/2,where Wact is the actual width of the fibril, Wapp is theapparent width of the fibril, H is the height of fibril, and Rt isthe radius of the tip [61]. Due to the large statistical range inthe measured apparent width, the standard deviationreported for the actual width is calculated using the equationfor Wact, the average apparent width, and the error in theheight. For all calculations, the radius of the tip, Rt, wasassumed to be 7 nm as provided by the manufacturer.Contour length and end to end distances were manuallymeasured using the freehand line scan option provided byAsylum research.

Fourier transform infrared spectroscopy

ATR-FTIR spectroscopy was performed on a BrukerOptik Vertex 70 (Ettlingen, Germany) spectrometerequipped with a BioATRcell II accessory (Harrick ScientificProducts, Inc.) as previously described [62]. Briefly, 30 μL


of freshly prepared monomer or pelletized aggregate wasplaced on a silicon crystal equilibrated to 24 °C. Samplespectra were acquired from 1000 to 3000 cm−1 with4 cm−1 resolution after background subtraction of thebuffer solution, both recorded for 1000 scans. Thespectra presented are an average of five such samplespectra. Combined peak and deconvolution analysiswas performed within the OPUS software analysispackage (Bruker Optik). After a horizontal baselinecorrection and smoothing algorithm (13–17 smoothingpoints), Gaussian curves were fitted from 1500 cm−1 to1700 cm−1 with fixed peaks identified by secondderivative analysis. As previously described [31], nativeβ-sheet bands were assigned from 1630 to 1641 cm−1;amyloid β-sheet bands, from 1611 to 1630 cm−1; turns,from 1663 to 1695 cm−1; and disorder, from 1647 to1654 cm−1.

Acknowledgements

We thank the Hud, Sulcheck, and Williams labora-tories for access to their CD spectropolarimeter, AFM,and fluorimeter, respectively. We also thank theUniversity of South Florida Department of Physicsand the Muschol laboratory for access to FTIR anddynamic light-scattering instruments. We also ac-knowledge assistance from Katherine Turnage forgenerating the MBP-OLF(TEV) construct, as well asPamela Chi, Dana Freeman, Elaine Nguyen, andLeigh Stafford in cell growth and purification. Thisworkwas funded by grants from theNational Institutesof Health (R01EY021205) and Pew Scholar inBiomedical Sciences program to R.L.L.

Appendix A. Supplementary data

Supplementary data to this article can be foundonline at http://dx.doi.org/10.1016/j.jmb.2013.12.002.

Received 30 July 2013;Received in revised form 6 November 2013;

Accepted 2 December 2013Available online 9 December 2013

Keywords:amyloid;

protein misfolding;protein structure;

circular dichroism;atomic force microscopy

Abbreviations used:TM, trabecular meshwork; ER, endoplasmic

reticulum; ThT, thioflavin T; AFM, atomic forcemicroscopy; ANS, anilinonaphthalene-1-sulfonate; FTIR,Fourier transform infrared spectroscopy; TEV, tobacco

etch virus.

References

[1] Stone EM, Fingert JH, Alward WL, Nguyen TD, Polansky JR,Sunden SL, et al. Identification of a gene that causes primaryopen angle glaucoma. Science 1997;275:668–70.

[2] Gobeil S, Rodrigue M-A, Moisan S, Nguyen TD, Polansky JR,Morissette J, et al. Intracellular sequestration of hetero-oligomers formed by wild-type and glaucoma-causing myocilinmutants. Invest Ophthalmol Visual Sci 2004;45:3560–7.

[3] Joe MK, Sohn S, Hur W, Moon Y, Choi YR, Kee C.Accumulation of mutant myocilins in ER leads to ER stressand potential cytotoxicity in human trabecular meshworkcells. Biochem Biophys Res Commun 2003;312:592–600.

[4] YamGH-F, Gaplovska-Kysela K, Zuber C, Roth J. Aggregatedmyocilin induces russell bodies and causes apoptosis:implications for the pathogenesis of myocilin-caused primaryopen-angle glaucoma. Am J Pathol 2007;170:100–9.

[5] Zhou Z, Vollrath D. A cellular assay distinguishes normal andmutant TIGR/myocilin protein. HumMolGenet 1999;8:2221–8.

[6] Suntharalingam A, Abisambra JF, O'Leary JC, Koren J,Zhang B, Joe MK, et al. Glucose-regulated protein 94 triageof mutant myocilin through endoplasmic reticulum-associateddegradation subverts a more efficient autophagic clearancemechanism. J Biol Chem 2012;287:40661–9.

[7] Polansky JR, Fauss DJ, Chen P, Chen H, Lutjen-Drecoll E,Johnson D, et al. Cellular pharmacology and molecularbiology of the trabecular meshwork inducible glucocorticoidresponse gene product. Ophthalmologica 1997;211:126–39.

[8] Orwig SD, Lieberman RL. Biophysical characterization of theolfactomedin domain of myocilin, an extracellular matrixprotein implicated in inherited forms of glaucoma. PLoS One2011;6:e16347.

[9] Gong G, Kosoko-Lasaki O, Haynatzki GR, Wilson MR.Genetic dissection of myocilin glaucoma. Hum Mol Genet2004;13:R91-102.

[10] Hewitt AW, Mackey DA, Craig JE. Myocilin allele-specificglaucoma phenotype database. Hum Mutat 2007;29:1–5.

[11] Burns JN, Orwig SD, Harris JL, Watkins JD, Vollrath D,Lieberman RL. Rescue of glaucoma-causing mutant myocilinthermal stability by chemical chaperones. ACS Chem Biol2010;5:477–87.

[12] Burns JN, Turnage KC, Walker CA, Lieberman RL. Thestability of myocilin olfactomedin domain variants providesnew insight into glaucoma as a protein misfolding disorder.Biochemistry 2011;50:5824–33.

[13] Donegan RK, Hill SE, Turnage KC, Orwig SD, Lieberman RL.The glaucoma-associated olfactomedin domain ofmyocilin is anovel calciumbinding protein. J Biol Chem 2012;287:43370–7.

[14] Orwig SD, Perry CW, Kim LY, Turnage KC, Zhang R,Vollrath D, et al. Amyloid fibril formation by the glaucoma-associated olfactomedin domain of myocilin. J Mol Biol2012;421:242–55.

[15] Eisenberg D, Jucker M. The amyloid state of proteins inhuman diseases. Cell 2012;148:1188–203.

[16] Zode GS, Kuehn MH, Nishimura DY, Searby CC, Mohan K,Grozdanic SD, et al. Reduction of ER stress via a chemicalchaperone prevents disease phenotypes in a mouse model ofprimary open angle glaucoma. J Clin Invest 2011;121:3542–53.

[17] Chiti F, Dobson CM. Protein misfolding, functional amyloid,and human disease. Annu Rev Biochem 2006;75:333–66.

[18] Kwon YH, Fingert JH, Kuehn MH, Alward WL. Primary open-angle glaucoma. N Engl J Med 2009;360:1113–24.

[19] Tomarev SI, Nakaya N. Olfactomedin domain-containingproteins: possible mechanisms of action and functions in

http://dx.doi.org/10.1016/j.jmb.2013.12.002


normal development and pathology. Mol Neurobiol2009;40:122–38.

[20] Gabelt BAT, Kaufman PL. Changes in aqueous humordynamics with age and glaucoma. Prog Retin Eye Res2005;24:612–37.

[21] Tamm ER, Ethier CR. Current aspects of aqueous humordynamics and glaucoma. Exp Eye Res 2009;88:618–9.

[22] Quigley HA. Number of people with glaucoma worldwide. BrJ Ophthalmol 1996;80:389–93.

[23] HoyerW,AntonyT,ChernyD,HeimG, JovinTM,SubramaniamV. Dependence of α-synuclein aggregate morphology onsolution conditions. J Mol Biol 2002;322:383–93.

[24] Kessel L, Johnson L, Arvidsson H, Larsen M. Therelationship between body and ambient temperature andcorneal temperature. Invest Ophthalmol Visual Sci2010;51:6593–7.

[25] Hider RC, Kupryszewski G, Rekowski P, Lammek B. Originof the positive 225-230 nm circular dichroism band inproteins. Its application to conformational analysis. BiophysChem 1988;31:45–51.

[26] Kelly SM, Jess TJ, Price NC. How to study proteins bycircular dichroism. Biochim Biophys Acta 2005;1751:119–39.

[27] Ranjbar B, Gill P. Circular dichroism techniques: biomolec-ular and nanostructural analyses- a review. Chem Biol DrugDes 2009;74:101–20.

[28] Buchner J, Renner M, Lilie H, Hinz HJ, Jaenicke R, KiefhabelT, et al. Alternatively folded states of an immunoglobulin.Biochemistry 1991;30:6922–9.

[29] Kelly SM, Price NC. The application of circular dichroism tostudies of protein folding and unfolding. Biochim BiophysActa 1997;1338:161–85.

[30] Hill SE, Miti T, Richmond T, Muschol M. Spatial extent ofcharge repulsion regulates assembly pathways for lysozymeamyloid fibrils. PLoS One 2011;6:e18171.

[31] Shivu B, Seshadri S, Li J, Oberg KA, Uversky VN, Fink AL.Distinct β-sheet structure in protein aggregates deter-mined by ATR-FTIR spectroscopy. Biochemistry2013;52:5176–83.

[32] Oliveberg M. Waltz, an exciting new move in amyloidprediction. Nat Methods 2010;7:187–8.

[33] Trovato A, Seno F, Tosatto SC. The PASTA server forprotein aggregation prediction. Protein Eng Des Sel2007;20:521–3.

[34] Tsolis AC, Papandreou NC, Iconomidou VA, HamodrakasSJ. A consensus method for the prediction of “aggregation-prone” peptides in globular proteins. PLoS One 2013;8:e54175.

[35] Ahmed AB, Kajava AV. Breaking the amyloidogenicity code:methods to predict amyloids from amino acid sequence.FEBS Lett 2013;587:1089–95.

[36] Wetzel R. Kinetics and thermodynamics of amyloid fibrilassembly. Acc Chem Res 2006;39:671–9.

[37] Uversky VN, Fink AL. Conformational constraints for amyloidfibrillation: the importance of being unfolded. BiochimBiophys Acta 2004;1698:131–53.

[38] Greenwald J, Riek R. On the possible amyloid origin ofprotein folds. J Mol Biol 2012;421:417–26.

[39] Chiti F, Dobson CM. Amyloid formation by globular proteinsunder native conditions. Nat Chem Biol 2009;5:15–22.

[40] Goldschmidt L, Teng PK, Riek R, Eisenberg D. Identifying theamylome, proteins capable of forming amyloid-like fibrils.Proc Natl Acad Sci U S A 2010;107:3487–92.

[41] Fitzpatrick AW, Debelouchina GT, Bayro MJ, Clare DK,Caporini MA, Bajaj VS, et al. Atomic structure and

hierarchical assembly of a cross-β amyloid fibril. Proc NatlAcad Sci U S A 2013;110:5468–73.

[42] Maurer-Stroh S, Debulpaep M, Kuemmerer N, Lopez de laPaz M, Martins IC, Reumers J, et al. Exploring the sequencedeterminants of amyloid structure using position-specificscoring matrices. Nat Methods 2010;7:237–42.

[43] Gosal WS, Morten IJ, Hewitt EW, Smith DA, Thomson NH,Radford SE. Competing pathways determine fibril morphologyin the self-assembly ofβ2-microglobulin into amyloid. JMol Biol2005;351:850–64.

[44] Kodali R, Williams AD, Chemuru S, Wetzel R. Aβ(1-40) formsfive distinct amyloid structures whose β-sheet contents andfibril stabilities are correlated. J Mol Biol 2010;401:503–17.

[45] Andersen CB, Hicks MR, Vetr i V, Vandahl B,Rahbek-Nielsen H, Thogersen H, et al. Glucagon fibrilpolymorphism reflects differences in protofilament backbonestructure. J Mol Biol 2010;397:932–46.

[46] Raman B, Chatani E, Kihara M, Ban T, Sakai M, HasegawaK, et al. Critical balance of electrostatic and hydrophobicinteractions is required for β 2-microglobulin amyloid fibrilgrowth and stability. Biochemistry 2005;44:1288–99.

[47] Goto Y, Calciano LJ, Fink AL. Acid-induced folding ofproteins. Proc Natl Acad Sci U S A 1990;87:573–7.

[48] Sasahara K, Yagi H, Sakai M, Naiki H, Goto Y. Amyloidnucleation triggered by agitation of β2-microglobulin under acidicand neutral pH conditions. Biochemistry 2008;47:2650–60.

[49] Sicorello A, Torrassa S, Soldi G,Gianni S, Travaglini-AllocatelliC, Taddei N, et al. Agitation and high ionic strength induceamyloidogenesis of a folded PDZ domain in native conditions.Biophys J 2009;96:2289–98.

[50] Shirahama T, Cohen AS. High-resolution electronmicroscopicanalysis of the amyloid fibril. J Cell Biol 1967;33:679–708.

[51] Malisauskas M, Zamotin V, Jass J, Noppe W, Dobson CM,Morozova-Roche LA. Amyloid protofilaments from thecalcium-binding protein equine lysozyme: formation of ringand linear structures depends on pH and metal ionconcentration. J Mol Biol 2003;330:879–90.

[52] Ding TT, Lee SJ, Rochet JC, Lansbury PT. Annular α-synuclein protofibrils are produced when spherical protofibrilsare incubated in solution or bound to brain-derived mem-branes. Biochemistry 2002;41:10209–17.

[53] Relini A, Torrassa S, Rolandi R, Gliozzi A, Rosano C, CanaleC, et al. Monitoring the process of HypF fibrillization andliposome permeabilization by protofibrils. J Mol Biol2004;338:943–57.

[54] Solomon JP, Page LJ, Balch WE, Kelly JW. Gelsolinamyloidosis: genetics, biochemistry, pathology and possiblestrategies for therapeutic intervention. Crit Rev Biochem MolBiol 2012;47:282–96.

[55] Solomon JP, Bourgault S, Powers ET, Kelly JW. Heparinbinds 8 kDa gelsolin cross-β-sheet oligomers and acceler-ates amyloidogenesis by hastening fibril extension. Bio-chemistry 2011;50:2486–98.

[56] Relini A, De Stefano S, Torrassa S, Cavalleri O, Rolandi R,Gliozzi A, et al. Heparin strongly enhances the formation ofβ2-microglobulin amyloid fibrils in the presence of type Icollagen. J Biol Chem 2008;283:4912–20.

[57] Perez M, Valpuesta JM, Medina M, Montejo de Garcini E,Avila J. Polymerization of tau into filaments in the presence ofheparin: the minimal sequence required for tau-tau interaction.J Neurochem 1996;67:1183–90.

[58] Pollard HB, Arispe N, Rojas E. Ion channel hypothesis forAlzheimer amyloid peptide neurotoxicity. Cell Mol Neurobiol1995;15:513–26.


[59] Tropea JE, Cherry S, Waugh DS. Expression and purificationof soluble His(6)-tagged TEV protease. Methods Mol Biol2009;498:297–307.

[60] Stelea SD, Keiderling TA. Pretransitional structural changesin the thermal denaturation of ribonuclease S and S protein.Biophys J 2002;83:2259–69.

[61] Fung SY, Keyes C, Duhamel J, Chen P. Concentration effecton the aggregation of a self-assembling oligopeptide. BiophysJ 2003;85:537–48.

[62] FoleyJ,HillSE,MitiT,MulajM,CieslaM,RobeelR,etal.Structuralfingerprints and their evolution during oligomeric vs. oligomer-free amyloid fibril growth. J Chem Phys 2013;139:121901.

The Glaucoma-associated Olfactomedin Domain of Myocilin Is a Novel Calcium Binding Protein

Documents

Transcript of The Glaucoma-associated Olfactomedin Domain of Myocilin Is a Novel Calcium Binding Protein