Apr 30, 2024
Липопротеины спинномозговой жидкости ингибируют α
Молекулярная нейродегенерация, том 18, номер статьи: 20 (2023) Цитировать эту статью 2110 Доступов 3 Цитирования 24 Подробности об альтметрических метриках Агрегация α-синуклеина (α-син) является характерной особенностью
Молекулярная нейродегенерация, том 18, номер статьи: 20 (2023) Цитировать эту статью
2110 Доступов
3 цитаты
24 Альтметрика
Подробности о метриках
Агрегация α-синуклеина (α-син) является характерной особенностью болезни Паркинсона (БП) и других синуклеинопатий. В настоящее время анализы амплификации семян α-сина (SAA) с использованием спинномозговой жидкости (СМЖ) представляют собой наиболее перспективные инструменты диагностики синуклеинопатий. Однако сам СМЖ содержит несколько соединений, которые могут модулировать агрегацию α-син в зависимости от пациента, потенциально подрывая неоптимизированные α-син-САА и препятствуя количественному определению семян.
В этом исследовании мы охарактеризовали ингибирующее влияние среды спинномозговой жидкости на обнаружение агрегатов α-син с помощью фракционирования спинномозговой жидкости, масс-спектрометрии, иммуноанализа, трансмиссионной электронной микроскопии, спектроскопии ядерного магнитного резонанса в растворе, высокоточного и стандартизированного диагностического SAA и различных условия агрегации in vitro для оценки спонтанной агрегации α-син.
Мы обнаружили, что высокомолекулярная фракция CSF (> 100 000 Да) сильно ингибирует агрегацию α-син, и определили, что липопротеины являются основными движущими силами этого эффекта. Прямое взаимодействие между липопротеинами и мономерным α-син не было обнаружено методом спектроскопии ядерного магнитного резонанса в растворе, однако с помощью трансмиссионной электронной микроскопии мы наблюдали комплексы липопротеин-α-син. Эти наблюдения согласуются с гипотезой о взаимодействии между липопротеинами и олигомерными/протофибриллярными интермедиатами α-син. Мы наблюдали значительно более медленную амплификацию семян α-сина в PD CSF, когда липопротеины добавлялись к реакционной смеси диагностических SAA. Кроме того, мы наблюдали снижение способности CSF ингибировать агрегацию α-син после иммунодеплеции ApoA1 и ApoE. Наконец, мы заметили, что уровни ApoA1 и ApoE в CSF значительно коррелировали с кинетическими параметрами SAA в n = 31 SAA-негативных контрольных образцах CSF, в которые были добавлены предварительно сформированные агрегаты α-син.
Наши результаты описывают новое взаимодействие между липопротеинами и агрегатами α-син, которое ингибирует образование α-син-фибрилл и может иметь соответствующие последствия. Действительно, донор-специфическое ингибирование CSF агрегации α-син объясняет отсутствие на сегодняшний день количественных результатов анализа кинетических параметров, полученных из SAA. Кроме того, наши данные показывают, что липопротеины являются основными ингибирующими компонентами спинномозговой жидкости, что позволяет предположить, что измерения концентрации липопротеинов могут быть включены в модели анализа данных, чтобы устранить мешающее влияние среды спинномозговой жидкости на попытки количественного определения α-син.
Болезнь Паркинсона (БП), деменция с тельцами Леви (ДЛБ) и множественная системная атрофия (МСА) представляют собой нейродегенеративные заболевания, патологически характеризующиеся наличием внутриклеточных включений α-син в уязвимых областях мозга и обычно называемые синуклеинопатиями. Анализы посевной амплификации (SAA), известные как циклическая амплификация неправильного сворачивания белка (PMCA) [1] и конверсия, индуцированная землетрясением в реальном времени (RT-QuIC) [2] в прионном поле, были недавно адаптированы для обнаружения агрегатов α-син. в биологической жидкости и тканях человека и может значительно улучшить диагностику синуклеинопатий в ближайшем будущем. SAA основаны на амплификации небольших количеств прионоподобных агрегатов α-син (зародышей α-син), присутствующих в биологических матрицах, которые размножаются in vitro путем рекрутирования добавленных рекомбинантных мономеров α-син посредством циклов элонгации и фрагментации [3, 4]. ]. Процесс амплификации контролируют с помощью тиофлавина-Т (ThT), флуоресцентного красителя, который с высоким сродством связывается с мотивами перекрестных β-листов амилоидных агрегатов. Примечательно, что SAA обнаружили семена α-сина в спинномозговой жидкости при продромальных случаях БП [5, 6], достигая такой же чувствительности, как и у пациентов с полномасштабным заболеванием [7,8,9]. Первоначальные отчеты показали корреляцию между скоростью агрегации и прогрессированием заболевания (оценка H&Y) [1] и уровнями синтетических агрегатов α-син, повышенных в спинномозговой жидкости [1, 7]. Однако эти результаты не были воспроизведены при анализе более крупных когорт [9, 10], а полуколичественный анализ с помощью серийных разведений также не коррелировал с прогрессированием заболевания [7, 9]. Было замечено, что спинномозговая жидкость как у здоровых людей из контрольной группы (HC), так и у больных синуклеинопатией ингибирует агрегацию α-синов по сравнению с единственным буфером [1, 11,12,13]. В результате протоколы SAA включают разбавление CSF для преодоления ингибирования и обеспечения эффективной амплификации семян α-сина [2, 11]. Этот эффект неоднократно наблюдался, но, что удивительно, еще не охарактеризован. Действительно, влияние CSF на агрегацию α-син может объяснить очевидное отсутствие корреляции между параметрами анализа с прогрессированием заболевания и нагрузкой α-син [9, 10].
2000, 0 < t1 < 100 h and t2 > 0. For some kinetic traces, a decrease in fluorescence was observed after reaching the second plateau. This known phenomenon is caused by the sequestration of ThT molecules by mature fibrils and by the sedimentation of HMW insoluble aggregates [21]. In these cases, the last descending part of the ThT profile was removed prior to fitting. Fitting was rejected when the adjusted determination coefficient R2 was below 0.3./p> 100 kDa), CSF constituents of MW between 100 and 50 kDa (100-50 kDa), CSF constituents of MW between 50 and 10 kDa (50-10 kDa), CSF constituents of MW between 10 and 3 kDa (10-3 kDa), and CSF constituents of MW below 3 kDa (< 3 kDa). We then analysed the inhibitory effect of each of these 6 fractions on the spontaneous aggregation of α-syn in the previously mentioned PBS conditions (i.e., those of Fig. 2B, C, E). There were clear differences in α-syn aggregation depending on the MW of the CSF fraction. Whole CSF and all the fractions with MW > 10 kDa drastically inhibited α-syn aggregation, while 10-3 kDa and < 3 kDa fractions showed comparable aggregation to the reaction without CSF components (PBS control) (Fig. 4B). We estimated the second fluorescence plateau (A2) using the double sigmoidal model and compared the results to the maximum fluorescence readings (Fmax) for all 6 CSF-derived samples (Fig. 4C). As expected, A2 and Fmax were very similar within each CSF derived sample, confirming the goodness of the fit, except for whole CSF and > 100 kDa for which fitting was not possible. Indeed, for whole CSF and > 100 kDa, fluorescence readings remained practically flat, impeding the fit through the double sigmoidal model. Suggesting lower inhibition of α-syn aggregation, A2 and Fmax were substantially higher in both 10–3 and < 3 kDa fractions, which were comparable to the PBS control. Some inhibition was observed in the < 3 kDa fraction, but this is probably the result of an increase in pH by air exposure during the fractionation procedure rather than an inhibition of α-syn aggregation by protein components (as described in Supplementary Material Solution NMR experiments on CSF fractions and Fig. S4-5). A2 and Fmax dropped significantly for 50–10 and 100–50 kDa fractions, consistent with a higher degree of α-syn aggregation inhibition, while whole CSF and > 100 kDa were the most inhibitory (Fig. 4C). Estimations of t2 show similar results, with whole CSF and > 100 kDa significantly slowing down aggregation (Supplementary Material Fig. S6). These results indicate that the main inhibitory components of CSF are enriched in the > 100 kDa fraction, which retained the same inhibitory effect as whole CSF. To identify protein candidates responsible for the inhibition of α-syn aggregation, we analysed the CSF derived samples by nLC-nESI HRMS/MS (Fig. 4D). We could not detect relevant amounts of proteins in the 10–3 kDa and < 3 kDa fractions by nLC-nESI HRMS/MS (just very low levels of albumin or albumin fragments in the 10-3 kDa fraction, see Supplementary Material, Table S2). In whole CSF, the most abundant of the ~ 300 proteins detected were albumin, transthyretin (TTR), apolipoproteins, and prostaglandin-D synthase (PGDS, also known as β-trace protein). Albumin and PGDS were found not particularly enriched in the > 100 kDa fraction, suggesting low inhibition effect of these two proteins on α-syn aggregation. Following the inhibitory effect on α-syn aggregation, apolipoproteins were more abundant in whole CSF and the > 100 kDa fraction. ApoA1 and ApoE were the most abundant apolipoproteins in the > 100 kDa fraction (~ 80%), while ApoJ and ApoD accounted for ~ 6% each. The higher abundance of fatty acids-rich lipoproteins and/or other HMW proteins in NPH1 vs NPH2, and the high abundance of lipoproteins in the most inhibitory fraction of CSF (> 100 kDa) suggest that these compounds in CSF could be the main driver of the inhibition of seeded and spontaneous aggregation of α-syn./p> 100 kDa fraction) and maximum fluorescence values (Fmax) estimated from individual ThT traces. Two scales of fluorescence intensity were used to better compare the results. Represented values correspond to the average of three replicates with error bars reflecting the SEM. D Relative concentration (emPAI score multiplied by protein molecular weight) of the most abundant protein constituents measured by nLC-nESI HRMS/MS. Apolipoproteins scores were summed together, with ApoA1 and ApoE being the most abundant (~ 85% of the total). Scores for fractions 10–3 and < 3 kDa are not shown since the protein content of these fractions was negligible with respect to the others/p> 100 kDa CSF fraction, we evaluated if HDL could retain the same level of inhibition when tested within a range of concentrations from 1 to 0.003 mg/mL (including the physiological ones of human CSF). Interestingly, we observed a dose-dependent partial inhibition of α-syn aggregation when adding 0.003 and 0.03 mg/mL, while 0.3 and 1 mg/mL completely blocked the formation of ThT-reactive aggregated species (Fig. 6A). The partial inhibition was most noticeable as a delay in the second inflection point (t2), although it was also observed as a reduction in fluorescence of the first plateau (A1) (Fig. 6A inset). ApoE and ApoA1 represent 50–60% of total CSF apolipoprotein, and their respective reported concentration in CSF is approximately 0.01 mg/mL and 0.004 mg/mL [26]. Considering 0.03 mg/mL as the physiological concentration of HDL in CSF, our results show that HDL partially inhibited α-syn aggregation at a physiological concentration and at a tenfold lower concentration. Although our experiments were performed with highly purified recombinant α-syn and other components were shown not to form ThT, OC, or A11 detectable aggregates, we used WB to detect monomeric recombinant α-syn as a secondary read-out for this experiment. In agreement with ThT readings, in the absence of HDL, there was a decrease in monomeric α-syn at the time of the first plateau (48 h) and most of the monomer was consumed by the time of the second plateau (180 h) (Fig. 6B, Supplementary Material Fig. S8A). However, we detected monomeric α-syn in the presence of HDL after 180 h of the reaction, with the monomer signal being found increased at increasing concentrations of HDL (highest at 1 mg/mL HDL and the lowest in absence of HDL, Fig. 6C). However, it is worth mentioning that, in the presence of 1 mg/mL HDL, we observed an additional band (around 150–200 kDa) of much lower intensity than the monomer band (Supplementary Material Fig. S8B-C), suggesting that high concentrations of HDL may have stabilised some prefibrillar oligomers, preventing their conversion into fibrils. Collectively, these results show that purified serum HDL (at physiological concentrations and in the absence of CSF milieu) is a potent inhibitor of α-syn aggregation, comparable to whole CSF and the > 100 kDa CSF fraction./p> 200 h t2). TTR did not inhibit α-syn spontaneous aggregation when tested within the physiological concentration range (0.03 and 0.3 mg/mL) (Fig. 7A-B). Nevertheless, TTR showed partial inhibition, most noticeable in Fmax/A2 when tested at 1 mg/mL, which has been reported to be near the plasma physiological concentration of TTR [29]. These results suggest that the size, density, or lipidic content of the lipoprotein are not critical for the inhibition of α-syn aggregation, although these factors may modulate the level of inhibition on α-syn aggregation./p> 100 kDa) retained the same inhibitory effect on α-syn aggregation as whole CSF, while fractions of smaller MW were less inhibitory. It should be considered that the pore size distribution of MWCO filters allows a low percentage of HMW components to pass through the filter in smaller MW fractions. This potential leak of HMW material into lower MW fractions could explain the lower but still detectable inhibitory effect observed in those fractions. Using MS we identified albumin, TTR, PGDS, and apolipoproteins to be the most abundant HMW components in the NC CSF pool used. The distribution of apolipoprotein in the CSF fractions correlated with the inhibition of each fraction on α-syn aggregation, while albumin, TTR, and PGDS did not. Apolipoproteins, particularly ApoA1 (28.3 kDa) and ApoE (34 kDa), were the most represented in the > 100 kDa fraction. Given the MW, these apolipoproteins were most likely assembled in lipoproteins in CSF. We then confirmed that serum-purified HDL in the absence of CSF milieu was able to inhibit α-syn spontaneous aggregation. Moreover, the partial inhibition was observed when evaluating HDL at CSF sub-physiological concentrations. These findings were confirmed by tracking the aggregation using dot blot with two different conformational antibodies and WB detecting the monomeric α-syn that was not converted into fibrils in the aggregation reaction. In these experiments, although most α-syn remained monomeric at 1 mg/mL HDL, we could spot the presence of a band at 150–200 kDa, suggesting that high HDL concentrations might have stabilised some prefibrillar oligomers, impeding their conversion into fibrils. Since CSF HDL are bigger in size than blood HDL and smaller than blood LDL [27, 28], we also evaluated the inhibitory effect of serum-purified LDL on α-syn aggregation and observed a similar or greater inhibitory effect than with HDL even at sub-physiological concentrations. We evaluated HDL and LDL at the same concentrations in terms of mg/mL, but because of the smaller molecular weight of HDL, molar concentration of HDL was greater than the concentration of LDL in our α-syn aggregation experiments. Thus, the higher inhibitory effect of LDL might indicate that the inhibition is driven by the lipidic fraction of the complex, although there are many reports showing that lipids promote aggregation of α-syn [31]./p> 100 kDa CSF fraction in PBS. The residues experiencing the largest decreases in signal intensity (smaller by one or more standard deviations with respect to the average value) are highlighted in light blue. The intensity ratios corresponding to overlapping peaks are highlighted in red (their values were not considered in the calculation of the average decreases and standard deviations). Fig. S5. CSF pH drift. The pH change due to the exposure of CSF to air was monitored over time in 500 μL of undiluted pooled CSF (A) and in the presence of PBS (400 μL CSF + 200 μL PBS 3x) in polypropylene vials with a Thermo Scientific Orion pH-meter equipped with a glass 6 mm diameter pHenomenal MIC 220 Micro electrode. Right before each measurement, the sample was vortexed for 20 sec and left open to air for another 20 sec. Fig. S6. Different CSF fractions differently affect α-syn aggregation. Mean fitted t2 parameters of samples with 40 μl of PBS/CSF fractions. The values displayed result from the average of three replicates with error bars reflecting the SEM. For whole CSF and the >100 kDa fraction the total duration of the experiment is shown due to the absence of appreciable aggregation. Fig. S7. Raw images of the dot-blot assays. A-B) Native images of the dot-blot assay performed on α-syn alone replicates at different timepoints with OC (A) and A11 (B) conformational antibodies. C-D) Native image of the dot-blot assay performed on the HSA and HDL containing samples with OC (C) and A11 (D) conformational antibodies. Fig. S8. WB experiments to track α-syn aggregation in the presence of human HDL. A) α-Syn aggregation patterns in samples collected at different timepoints of the spontaneous aggregation process, was monitored by WB using Syn211 antibody (4-20% SDS-PAGE, 2 μg protein loaded). Monomeric α-syn decreases as t increases due to the formation of fibrils. B) In a similar way, a WB with Syn211 was performed on the reaction products obtained after 180 h, at different HDL concentrations with and without α-syn (exposure time 210 s). C) The experiment was then repeated by doubling the amount of sample loaded into the gel to better highlight the presence of oligomeric species (exposure time 30 s). Under these conditions, chosen to better visualize the signal at 150-200 kDa, the α-syn monomer bands at 1 and 0.3 mg/mL HDL may not quantitively reflect monomer concentration (overloaded lanes). Fig. S9. Representative TEM images of α-syn incubated with CSF. Representative TEM images obtained by analyzing samples obtained by the co-incubation of α-syn 0.7 mg/mL at 37 °C with pooled human CSF (1:5 ratio with respect to total reaction volume). Samples were subjected to cycles of incubation (13 min) and shaking (double-orbital, 2 min) at 500 rpm. Fig. S10. NMR titrations of α-syn with HDL, LDL and TTR. A) Intensity decreases of the signals of two-dimensional (2D) 15N–1H HSQC experiments acquired at 950 MHz at T = 283 K on 15N labelled α-syn (100 μM) in PBS after the addition of 0.57 mg/mL serum-derived HDL. The intensity ratios corresponding to overlapping peaks are highlighted in red. B) Intensity decreases of the signals of two-dimensional (2D) 15N–1H HSQC experiments acquired at 950 MHz at T = 283 K on 15N labelled α-syn (100 μM) in PBS after the addition of 1 mg/mL serum-derived LDL. C) Intensity decreases of the signals of two-dimensional (2D) 15N–1H HSQC experiments acquired at 950 MHz at T = 283 K on 15N labelled α-syn (100 μM) in PBS after the addition of 3 mg/mL TTR. Fig. S11. WB experiments performed on immunodepleted CSF. WB experiments were performed with anti-ApoA1 (MIA1404) and anti-ApoE (PA5-27088) antibodies on neat CSF and supernatants (400 μL CSF, 100 μL slurry) resulting from immunoprecipitation procedure performed using different quantities of the same antibodies (IP CSF samples), immunoprecipitation performed without antibodies (CSF IP no Ab). The conditions relative to IP CSF ApoA1 60 µg and IP CSF ApoE 20 µg were then selected for Protein aggregation assays. Table S1. Concentration factors and final volumes of the CSF fractions./p>
Русский
English
Français
Deutsch
Español
Italiano
Português
日本語
한국어
Дом