TECHNICAL REPORTS
Detection of Survival Motor Neuron Protein in Buccal Cells Through Electrochemiluminescence-Based Assay Hannes Steinkellner,1,* Julia Etzler,1,* Bernhard M. Gmeiner,2 and Franco Laccone1 Institutes of 1Medical Genetics and 2Medical Chemistry, Medical University of Vienna, Vienna, Austria. *These two authors contributed equally to this work.
ABSTRACT Spinal muscular atrophy (SMA) is a severe autosomal recessive disorder affecting one in every 10,000 live births. The disease is characterized by loss of alpha-motor neurons in the spinal cord that leads to progressive atrophy and weakness of limb and trunk muscles. This neuromuscular disorder results from deletions and/ or mutations within the survival motor neuron 1 (SMN1) gene, leading to a pathologically decreased expression of functional full-length SMN protein. Here we report on the investigation to measure SMN protein levels through electrochemiluminescence immunoassay (ECLIA). This simple assay is a highly quantitative method able to measure SMN protein levels in human, mouse, and rat samples throughout a wide working range with low intra- and interassay error. The sensitivity for human SMN is 30 pg/mL and provides a new tool for the set up of highthroughput screening for basic research. Moreover, we describe a novel tool for a noninvasive assessment of SMN in buccal cells derived from healthy donors, SMA carriers, and SMA patients. The availability of a validated quantitative ECLIA should improve the investigation of novel compounds for the treatment of SMA.
INTRODUCTION
S
pinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disorder characterized by hypotonia, progressive proximal muscle weakness, and respiratory distress that result from sequential degeneration of alpha-motor neurons in the anterior horn of the spinal cord.1 Although carrier rates vary considerably among different ethnic groups,2 SMA is one of the most common monogenetic disorders, with an incidence of 1 in 6,000 to 1 in 10,000 live births,3,4 and a leading genetic cause
DOI: 10.1089/adt.2015.635
of infant mortality. Based upon the severity of its clinical manifestations, namely the age of disease onset and the motor milestones achieved, the spectrum of SMA can be subdivided into four types.5 In the most severely affected patients (Type I), symptoms manifest by 6 months of age and usually lead to death by respiratory failure before age 2, whereas Type II and III patients present with first symptoms after the age of 6 months and develop the ability to sit (Type II) or walk (Type III), respectively. Comparatively, mildly affected patients with a disease onset after the age of 30 and a normal life expectancy are classified as Type IV SMA. In 1995, the survival motor neuron 1 (SMN1) gene was identified as an SMA-determining gene,1 and it has since been shown that homozygous deletion or mutational inactivation of SMN1 is causative of all four types of SMA. The broad clinical spectrum of the disorder can, in part, be explained by its unique genetic background: the SMN1 gene is located within a duplicated and inverted 500 kb chromosomal segment on 5q13 that comprises a neighboring nearly identical gene called SMN2. The two genes differ in five base pairs only, all of which are located toward their 30 ends,1,6 and they are both ubiquitously expressed. Full-length transcripts derived from both SMN1 and SMN2 encode an identical 294 AA product, the SMN protein. However, due to a critical C-to-T transition at position + 6 of exon 7 of the SMN2 gene, only about 10% of its transcripts are correctly spliced full-length mRNAs, while the vast majority of transcripts derived from SMN2 lack exon 7,5,7,8 resulting in the translation of a truncated SMND7 protein that is highly unstable and cannot functionally replace SMN.9,10 Although the homozygous inactivation of SMN1 in SMA patients cannot be fully compensated for, a small amount of functional SMN protein is still produced by the SMN2 gene. Consequentially, an increased number of SMN2 copies is thought to moderate the lack of SMN protein, thus reducing SMA severity, which explains why SMN2 copy number has been shown to inversely correlate with the clinical severity of the disorder.11–13 In SMA patients, the SMN2 copy number can reach up to eight and usually varies between 1 and 4
ª MARY ANN LIEBERT, INC. VOL. 13
NO. 3 APRIL 2015 ASSAY and Drug Development Technologies
167
STEINKELLNER ET AL.
copies,12 whereas a complete loss of SMN2 in an SMN1 null genome is embryonically lethal in humans14—in other words, every SMA patient possesses at least one copy of SMN2 and is thus expected to produce a basic level of SMN protein. Given the particular genetics that gives rise to SMA, enhancing the production of full-length SMN protein from the SMN2 gene, which is present in all patients, has become a major therapeutic approach.15,16 For any drug discovery program based on that mode of action, a key prerequisite is the reliable and accurate quantification of SMN protein levels in various fluids, cells, and tissues. Moreover, as some of the preclinical drug candidates eventually enter clinical testing, these quantification methods also need to perform rapidly and on a high-throughput basis. To meet these criteria, different SMN-specific enzyme-linked immunosorbent assays (ELISAs) have been developed during the past years that allow for both SMA drug screening and SMN measurement in human and mouse tissue samples, including peripheral blood mononuclear cells (PBMCs).16–19 Although ELISAs can detect much lower protein quantities than classical western blotting (dynamic range of 50– 3,200 pg/mL),19 in terms of sensitivity they are outperformed by a recently developed detection method called electrochemiluminescence immunoassay (ECLIA) whose principle of function is described elsewhere.20,21 The aim of this study was the development and verification of this high-throughput assay for simple and accurate quantification of SMN protein levels that allows a highly sensitive detection limited to 30 pg/mL.
MATERIALS AND METHODS All the samples used in this study were taken after the participants provided their consent by signing a statement of agreement. The SMA carrier provided written informed consent before providing buccal cells used in this study. For the SMA patient, we obtained written informed consent from his guardian.
with embryonic rat dorsal root ganglion neurons),22 human Schwann cells (hSCs, kindly provided by Prof. Ahmet Ho¨ke, John Hopkins University),23 and an SMA fibroblast cell line (#GM22592; Corriel Cell Repositories). The cells were maintained in Dulbecco’s modified Eagle’s medium, including high glucose and stabilized glutamine (Invitrogen), supplemented with 10% fetal bovine serum (FBS, from PAA Laboratories; for hSCs, we used FBS from Sigma-Aldrich) and penicillin/ streptomycin (Gibco/Life Technologies). The hSC medium was further enriched with 0.2% glucose (Gibco) and 2 mM forskolin (Sigma-Aldrich). Cells were incubated in a humidified atmosphere (5% CO2, 37C) for 48 h before sample preparation. Sample Preparation Cells obtained from the culture of different sources were harvested and washed twice with phosphate-buffered saline (PBS, pH 7.4), centrifuged at 250 g for 10 min, and stored at - 80C for further use. In addition, human lymphocytes were obtained from isolated fresh blood samples with LymphoprepTM (Axis-Shield) according to the manufacturer’s procedure. The cell pellets were dissolved in ice-cold lysis reagent (1 · tris-buffered saline pH 7.4, 1% triton X-100, 1 mM ethylenediaminetetraacetic acid, 1 mM ethylene glycol tetraacetic acid, 0.05% sodium dodecyl sulfate, 0.1% sodium deoxycholate, and 2% nonidet-P40) with freshly added protease inhibitors (1 mM phenylmethylsulfonyl fluoride [from Sigma], 5 mM sodium fluoride [Merck Chemicals], 1 mM sodium orthovanadate, and protease inhibitor cocktail [#PI8340 from Sigma Aldrich]) and incubated on ice for about 25 min before centrifugation at 10,000 g for 12 min at 4C. Then, the supernatant was analyzed by the BCA Protein Assay Kit (Thermo Scientific Pierce) for protein concentration.
Antibodies We used a primary mouse SMN monoclonal antibody as the capture antibody from Sigma-Aldrich (clone 2B1, #S2944) and a secondary SMN rabbit polyclonal antibody from Proteintech (#11708-1-AP); the goat anti-rabbit HRP-Sulfo TAGTM antibody from Meso Scale Discovery (MSD) was applied as a detection antibody. The recombinant human SMN1 standard protein was purchased from Enzo Life Sciences (#ADI-NBP-201-050).
Buccal Cell Preparation Buccal cells were harvested using MasterAmp Buccal Swab Brushes (Epicentre Biotechnology) by firmly pressing a brush against the inner cheek and swabbed while twirling for 60 s on each cheek. Both brushes were then immersed in the same tube containing 500 mL of ice-cold PBS and twirled gently to dislodge the cells. Samples were then microcentrifuged at 10,000 g for 10 min at 4C, and the cell pellet was lysed in the ice-cold lysis reagent as described in the sample preparation section. The protein concentration was calculated using the BCA protein assay as described above.
Cell Culture The cell lines used for the experiments were F11 (a fusion product of the mouse neuroblastoma cell line N18TG-2
SMN ECLIA Protocol The capture antibody was coated in the 96-well Multi-Array High Bind plates (MSD) by overnight incubation at 4C. Next
168 ASSAY and Drug Development Technologies APRIL 2015
ª MARY ANN LIEBERT, INC.
SMN PROTEIN AND ECLIA ASSAY
day, the plates were blocked at room temperature with 125 mL MSD Blocker A. The plates were washed thrice with PBSTween (0.05%) before adding the samples (25 mL per well diluted in 1% MSD Blocker or cell lysis buffer). The sample preparations were incubated at room temperature for another 90 min or at 4C for 16 h in case of buccal cell lysates. Each well was incubated with the polyclonal rabbit anti-SMN antibody (0.5 mg/mL in 1% MSD Blocker A) for 75 min followed by one incubation of 0.75 mg/mL of MSD Sulfo-TAG goat anti-rabbit detection antibody in 1% MSD Blocker A. One hundred fifty microliter Tris-based Read Buffer T (1 · ) with surfactant (MSD) was added to the plate and signals from ECL were captured by the Sector Imager 2400 reader (MSD) and recorded as signal counts (Table 1). SMN Copy Number Quantification of SMN1 and SMN2 copy number in four healthy donors was performed by Multiplex Ligation-
Table 1. Spinal Muscular Atrophy Electrochemiluminescence Immunoassay Protocol Table Step
Parameter
Value
Description
Fig. 1. Standard curve of recombinant human SMN1. Error bars show standard error at each dilution (n = 3). Linear regression in a range from 0.76 to 5,000 pg per well was generated using GraphPad Prism software. SMN, survival motor neuron.
dependent Probe Amplification (MLPA) using P021-A2 and P060-B2 SALSA MLPA probemix (MRC-Holland) according to the manufacturer’s instructions. Genomic DNA was extracted from blood lymphocytes by standard salt precipitation,24 and a total of 150 ng was used in each reaction. MLPA was run in duplicate for each donor and four different DNA standards.
1
Capture antibody
25 mL
1:4,000 in PBS buffer
2
Incubation
Overnight
Seal the plate, 4C
3
Blocking
125 mL
MSD Blocker A in PBS buffer
4
Incubation
90 min
Seal the plate, RT, shaking
5
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
6
Standard/samples
25 mL
In MSD Blocker A or Lysis buffer
7
Incubation
90 min
Seal the plate, RT, shaking
8
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
Table 2. Working Ranges of Different Sample Materials
9
Detection antibody
25 mL
1:3,330 in PBS buffer
Sample type
10
Incubation
75 min
Seal the plate, RT, shaking
HepG2, HEK 293
11
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
Human fibroblasts: SMA patients/control
1–5
12
Sulfo-TAG
25 mL
1:500 in 1% MSD Blocker A in PBS buffer
Murine fibroblasts
1–5
Buccal cells: patients/control
1–10
13
Incubation
60 min
Seal the plate, RT, shaking
14
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
15
Read Buffer T
150 mL
1:4 in Aqua Dest
16
Assay readout
CCD camera
MSD Sector Imager 2400
Statistical Analysis Statistical analysis of the measurement data was performed with GraphPad Prism software. Results are presented as mean – standard deviation and mean – standard error of the means. Statistical significance of differences was examined by using
PBS, phosphate-buffered saline; RT, room temperature; MSD, Meso Scale Discovery.
ª MARY ANN LIEBERT, INC. VOL. 13
Working range [lg] 0.1–2
F11 Rat DRG cells
0.5–2
Human Schwann cells
0.1–2
Mouse brain lysates
1–4
Suggested working ranges for various human, mouse, and rat sample materials. SMA, spinal muscular atrophy; DRG, dorsal root ganglion.
NO. 3 APRIL 2015
ASSAY and Drug Development Technologies 169
STEINKELLNER ET AL.
Table 3. Intra-Assay and Interassay Precision of Survival Motor Neuron Electrochemiluminescence Immunoassay pg SMN/lg Protein Well 1
Well 2
Well 3
Mean
SD
SEM
%CV
F11—rat DRG cells
1390.14
1425.14
1405.29
1406.86
17.55
10.13
1.25
Human Schwann cells
2223.10
2209.8l
2223.10
2245.35
38.54
22.25
1.72
PBMCs
8.57.32
891.78
846.49
865.20
23.65
13.65
2.73
Intra-assay
in the reactivity of the assay toward the SMN protein of different species. The measurement of different samples on three ensuing days yields an intraday precision of coefficient of variation (CV) £ 2.73% and an interday precision of CV £ 7.09% (Table 3).
Quantification of SMN Protein Levels in Human Buccal Cells To investigate whether our ECLIA would SMA fibroblasts 855.51 859.30 841.62 852.15 9.31 5.38 1.09 be sensitive enough to quantify the SMN Interassay protein in human buccal cells, we isolated buccal cells from one male healthy F11—rat DRG cells 1367.09 55.94 32.30 4.09 adult. The protocol adapted to this was as Human Schwann cells 2161.07 121 99 70.43 5.54 described in the Materials and Methods section because higher accuracy was PBMCs 702.54 23.85 16.86 3.39 achieved by overnight incubation of the SMA fibroblasts 835.93 59.24 34.20 7.09 samples. Intra-assay precision (n = 3 wells per day) and interassay precision (n = 3 separate days). The linearity over a wide range of SD, standard deviation; SEM, standard error mean; CV, coefficient of variation. buccal cell lysates from a healthy donor prompted us to investigate SMN protein levels in a small cohort of healthy donors (n = 4, Fig. 2). For the Student’s t-test. Significant differences are marked in the this, we collected buccal cells from an unmatched group of figures with *(P < 0.05), **(P < 0.01), and ***(P < 0.001). controls thrice within a 48-h interval (Table 4). Although we found a high variability in SMN protein levels RESULTS (Fig. 3) in this group of healthy donors, we investigated the Assay Specificity, Reproducibility, and Sensitivity possibility to measure SMN protein levels of buccal cell lysates The ECLIA is based on 96-microwell plates with carbon in an SMA carrier and patient. We collected buccal cells from electrodes built into the bottom for ultrasensitive detec25 one SMA carrier (male, 36 years old) and one SMA Type III tion. A standard curve with recombinant human SMN1 is shown in Figure 1. Recombinant human SMN1 can be measured over a wide range (0.76–5,000 pg per well with R2 = 0.9983) with high accuracy, which equals a dynamic range of 30–200,000 pg/mL. The recovery rate was estimated by spiking a definite amount of different cell lysates with low and high amounts of recombinant human SMN1. The recovery rate of spiked recombinant SMN1 with human SMA fibroblasts was 80% (high amounts) and 90% (low amounts). Similarly, we estimated a recovery rate for F11 cells of 79% (high amounts) and 93% (low amounts) (Data not shown). Validation of the Assay The SMN ECLIA is able to quantify SMN protein levels in a variety of cell lines and tissues of human, mouse, and rat origin. The working range was found to vary among different sample types (Table 2), which on one hand is most likely proportional to the concentrations of SMN per mg protein in different tissues and on the other hand also reflects differences
170 ASSAY and Drug Development Technologies APRIL 2015
Fig. 2. Linearity of buccal cell lysate. The assay remains sensitive and in the linear range (R2 = 0.98) from 0.3 to 4 mL buccal cell lysate per well.
ª MARY ANN LIEBERT, INC.
SMN PROTEIN AND ECLIA ASSAY
Table 4. Demographic Values of Healthy Donors, One Spinal Muscular Atrophy Carrier, and One Spinal Muscular Atrophy Patient Control (No.)
Sex
Age
Copies (SMN1/SMN2)
1
Female
29
2/2
2
Male
31
2/2
3
Male
61
2/2
4
Male
55
2/3
SMA carrier
Male
35
1/3
SMA patient
Male
6
-/3
Individual values of age (years) at the study, sex, and copy numbers of SMN1 and SMN2 are shown. No genotyping was performed on SMA carriers and SMA patients in this study, and the fathers self-identified themselves as SMA carriers and patients based on prior testing and diagnoses. SMA, spinal muscular atrophy.
patient (male, 6 years old) and measured SMN protein levels from buccal cell lysates with the ECLIA-based assay. As shown in Figure 4, SMN was readily detected in buccal cell lysates from the SMA carrier and a Type III SMA patient. The SMN protein level in the SMA carrier was slightly reduced compared to a sex- and age-matched healthy donor (Ctrl 2). The amount of SMN protein in the SMA patient was highly increased in relation to both, the control and the carrier, but could not be compared to an age-matched healthy control in this preliminary assessment.
Fig. 3. SMN protein levels in human buccal cells from healthy adults. Buccal cells were collected preprandial thrice from four healthy adults (Ctrl 1–4) on five consecutive days at every 48 h. SMN protein levels were measured in buccal cell extracts and normalized to protein levels. The data presented are mean – SD of triplicate wells (n = 3). Ctrl, control; SD, standard deviation.
Fig. 4. SMN protein levels in human buccal cells from SMA carrier and SMA patient. SMN protein levels in human buccal cells derived from SMA carrier and SMA patient compared to healthy control (Ctrl 2). The data presented are mean – SD of triplicate wells (n = 3). SMA, spinal muscular atrophy.
DISCUSSION We developed and validated an ECLIA for the quantitative measurement of SMN protein levels in human, mouse, and rat samples. Human SMN, which is ubiquitously expressed, has a molecular weight of 38 kDa and localizes to both the cytoplasm and the nucleus. It is part of a multiprotein complex that contains multiple SMN molecules and a number of stoichiometrically interacting proteins, Gemins 2–8.12 As with other SMN quantification methods, lysis reagents used in the ECLIA protocol have to be chosen carefully to ensure extraction and dissociation of both cytoplasmic and nuclear SMN.16 In the protocol described here, a commercially available lysis buffer known to work well in the ECLIA setting (MSD) is used, slightly modified for SMN extraction. Compared to other extraction buffers used in preliminary trials, it yielded higher concentrations of SMN in relation to whole protein (data not shown), suggesting a more efficient extraction of nuclear SMN. Both antibodies adapted for our ECLIA protocol are used in a well-established ELISA that has recently been used for measurement of blood SMN levels in a large clinical SMA study.13,19 The primary antibody (2B1) is a mouse monoclonal antibody that binds to the N-terminus of human SMN at amino acid positions 14–2019,26; the epitope for the secondary antibody (11708-1-AP), which is a rabbit polyclonal antibody, is located between amino acids 197 and 204 of human SMN.19 As both epitopes are expected to be present in full-length SMN as well as in SMND7, it is conceivable that both proteins are detected by our ECLIA, thus potentially interfering with the measurement of slight shifts in the ratio of SMN/SMND7.
ª MARY ANN LIEBERT, INC. VOL. 13 NO. 3 APRIL 2015
ASSAY and Drug Development Technologies 171
STEINKELLNER ET AL.
This should be kept in mind when using the ECLIA in preclinical screenings for drugs that increase the rate of exon7inclusion in SMN2-derived transcripts. Within the context of clinical studies, however, SMND7 is unlikely to substantially distort the quantification of SMN in tissue samples, as the truncated protein contains a potent degradation signal that causes it to be rapidly degraded.10 With regard to the assessment of SMN levels in murine samples using ELISA, concerns have been raised by Kobayashi et al. as to the differences in sensitivity. The authors report on a marked difference in reactivity of the assay to the SMN protein in humans and mice, respectively, with recombinant mouse SMN being detected at only about 10% the levels of recombinant human SMN.19 It is reasonable to assume that this holds true for the ECLIA as well, as the two assays are based on the exact same primary and secondary antibody. The ECLIA described here has a broad dynamic range (0.76– 5,000 pg/well with R2 = 0.9983) and a detection limit of 30 pg/ mL for the human SMN protein. It is therefore more sensitive than sandwich format ELISAs, whose reported detection limits range from 25 ng/mL18 to 50 pg/mL.19 As a result, SMN levels of various origin can be reliably assessed even at great dilutions, which means the ECLIA is less prone to matrix effects than other protein quantification methods and thus less subject to intersample variation. In line with its enhanced sensitivity, the ECLIA readily detected SMN in buccal swabs of a Type III SMA patient, an adult SMA carrier, and a healthy adult donor. To our knowledge, this is the first report of SMN protein quantification in human buccal cells. As they are particularly easy to access, it is tempting to speculate that buccal cells might constitute a noninvasive and practicable way to monitor SMN levels in future trials of SMNupregulating therapeutics. However, earlier studies using ELISA to assess SMN levels in different human tissue types have pointed at various pitfalls with regard to that approach. First, quantitative evaluation of murine tissues suggest that Smn expression varies widely across tissues of the same genotype, with Smn levels being found to be lowest in the primary tissues associated with SMA, that is, nerve, spinal cord, and muscle.19 It has been pointed out that SMN levels measured in PBMCs do not necessarily fully reflect SMN expression in more disease-relevant cells like motor neurons or muscle cells.13 The same holds true for buccal cells and any other peripheral tissues. Second, age is an important factor to consider when assessing SMN levels in clinical specimens, as there is evidence of a general decline of SMN levels across tissue types and species,16 and several studies using ECLIA to measure SMN levels in PBMCs of SMA patients, carriers, and controls have
172 ASSAY and Drug Development Technologies APRIL 2015
stressed the necessity to control for age when performing any statistical analysis.13,19 Third, even during a short period of time, SMN levels measured in PBMCs fluctuate substantially within individuals; when assessed at five time points over the course of 1 month, blood SMN levels were shown to vary from 2 to 10 times in six different healthy adults.19 Whether SMN levels measured in buccal cells using ECLIA are subject to such an extent of intraindividual fluctuation has to be evaluated by repeated within-subject testing, preferably normalizing SMN signals by cell counts instead of total protein. Finally, the finding that there is no clear correspondence between SMN protein levels in blood cells and disease severity raises fundamental concerns with regard to using peripheral SMN protein levels as an indicator of response to SMA treatment; while a correlation of protein quantity with phenotype could be established for severely affected Type I and Type II patients, no substantial decrease in the SMN protein is detectable in Type III patients.27,28 In line with this observation, a recent multicenter clinical trial using SMN ELISA to assess SMN levels in PBMCs of adult Type III SMA patients did not show any correlation between protein quantity and motor performance, leading the authors to suggest that SMN protein levels in blood cells are not suitable as biomarkers in clinical trials on SMA.29 In light of the data acquired from Type III SMA blood cells, our preliminary finding of a high SMN protein level in buccal cells of one Type III SMA patient is not that surprising, given the age difference between the patient (6 years) and the healthy control (31 years). It is doubtful that SMN levels in buccal cells will provide a more reliable indicator of disease severity in a mildly affected SMA patient than PBMCs, thus compromising the use as a biomarker in SMN upregulation trials. Taken together, the use of ECLIA to monitor SMN levels in buccal cells as a biomarker in clinical SMA drug trials is far from being straightforward. However, our data suggest that it can be effectively used to measure SMN protein levels as well as slight changes thereof in cell models of different species with high accuracy and specificity over a wide range. In terms of sensitivity, the ECLIA outperforms previous quantification methods. Furthermore, due to its relatively easy handling in a 96-well plate format, the assay may represent a novel tool of great value for high-throughput drug discovery screenings.
ACKNOWLEDGMENTS We gratefully thank Dr. Mark Du¨rr from Meso Scale Discovery (MSD, Germany) for his help with the ECLIA optimization and Dr. Oliver Brandau for collection of blood samples. Furthermore, we thank all donors for buccal cell samples.
ª MARY ANN LIEBERT, INC.
SMN PROTEIN AND ECLIA ASSAY
AUTHOR CONTRIBUTIONS All the authors of this article fulfilled the criteria of authorship. H.S. performed the experiments. H.S., J.E., B.M.G., and F.L. designed the study and wrote the article.
DISCLOSURE STATEMENT No competing financial interests exist.
REFERENCES 1. Lefebvre S, Burglen L, Reboullet S, et al.: Identification and characterization of a spinal muscular atrophy-determining gene. Cell 1995;80:155–165. 2. Hendrickson BC, Donohoe C, Akmaev VR, et al.: Differences in SMN1 allele frequencies among ethnic groups within North America. J Med Genet 2009; 46:641–644. 3. Pearn J: Incidence, prevalence, and gene frequency studies of chronic childhood spinal muscular atrophy. J Med Genet 1978;15:409–413. 4. Crawford TO, Pardo CA: The neurobiology of childhood spinal muscular atrophy. Neurobiol Dis 1996;3:97–110. 5. Wirth B, Brichta L, Hahnen E: Spinal muscular atrophy: from gene to therapy. Semin Pediatr Neurol 2006;13:121–131. 6. Burglen L, Lefebvre S, Clermont O, et al.: Structure and organization of the human survival motor neurone (SMN) gene. Genomics 1996;32:479–482. 7. Monani UR, Lorson CL, Parsons DW, et al.: A single nucleotide difference that alters splicing patterns distinguishes the SMA gene SMN1 from the copy gene SMN2. Hum Mol Genet 1999;8:1177–1183. 8. Lorson CL, Hahnen E, Androphy EJ, Wirth B: A single nucleotide in the SMN gene regulates splicing and is responsible for spinal muscular atrophy. Proc Natl Acad Sci U S A 1999;96:6307–6311. 9. Lorson CL, Androphy EJ: An exonic enhancer is required for inclusion of an essential exon in the SMA-determining gene SMN. Hum Mol Genet 2000;9: 259–265. 10. Cho S, Dreyfuss G: A degron created by SMN2 exon 7 skipping is a principal contributor to spinal muscular atrophy severity. Genes Dev 2010;24:438–442. 11. Feldkotter M, Schwarzer V, Wirth R, Wienker TF, Wirth B: Quantitative analyses of SMN1 and SMN2 based on real-time lightCycler PCR: fast and highly reliable carrier testing and prediction of severity of spinal muscular atrophy. Am J Hum Genet 2002;70:358–368. 12. Wirth B, Brichta L, Schrank B, et al.: Mildly affected patients with spinal muscular atrophy are partially protected by an increased SMN2 copy number. Hum Genet 2006;119:422–428. 13. Crawford TO, Paushkin SV, Kobayashi DT, et al.: Evaluation of SMN protein, transcript, and copy number in the biomarkers for spinal muscular atrophy (BforSMA) clinical study. PLoS One 2012;7:e33572. 14. Passini MA, Cheng SH: Prospects for the gene therapy of spinal muscular atrophy. Trends Mol Med 2011;17:259–265. 15. Lorson CL, Rindt H, Shababi M: Spinal muscular atrophy: mechanisms and therapeutic strategies. Hum Mol Genet 2010;19:R111–R118. 16. Cherry JJ, Kobayashi DT, Lynes MM, et al.: Assays for the identification and prioritization of drug candidates for spinal muscular atrophy. Assay Drug Dev Technol 2014;12:315–341. 17. Nguyen thi M, Humphrey E, Lam LT, et al.: A two-site ELISA can quantify upregulation of SMN protein by drugs for spinal muscular atrophy. Neurology 2008;71:1757–1763. 18. Piepers S, Cobben JM, Sodaar P, et al.: Quantification of SMN protein in leucocytes from spinal muscular atrophy patients: effects of treatment with valproic acid. J Neurol Neurosurg Psychiatry 2011;82:850–852.
19. Kobayashi DT, Olson RJ, Sly L, et al.: Utility of survival motor neuron ELISA for spinal muscular atrophy clinical and preclinical analyses. PLoS One 2011;6: e24269. 20. Yazuburo Namba MU, Osamu Suzuki: Highly sensitive electrochemiluminescence immuneassay using the ruthenium chelate-labeled antibody bound on the magnetic microbeads. Anal Sci 1999;15:1087–1093. 21. Debad JB, Glezer EN, Leland JK, Wohlstadter J, Sigal GB: Clinical and biological applications of ECL. In: Electrogenerated Chemiluminescence. Bard AJ (ed.), pp. 359–396. Marcel Dekker, New York, 2004. 22. Fan SF, Shen KF, Scheideler MA, Crain SM: F11 neuroblastoma · DRG neuron hybrid cells express inhibitory mu- and delta-opioid receptors which increase voltage-dependent K + currents upon activation. Brain Res 1992;590:329–333. 23. Lehmann HC, Chen W, Mi R, et al.: Human Schwann cells retain essential phenotype characteristics after immortalization. Stem Cells Dev 2012;21:423– 431. 24. Miller SA, Dykes DD, Polesky HF: A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16:1215. 25. Steinkellner H, Scheiber-Mojdehkar B, Goldenberg H, Sturm B: A high throughput electrochemiluminescence assay for the quantification of frataxin protein levels. Anal Chim Acta 2010;659:129–132. 26. Liu Q, Dreyfuss G: A novel nuclear structure containing the survival of motor neurons protein. EMBO J 1996;15:3555–3565. 27. Lefebvre S, Burlet P, Liu Q, et al.: Correlation between severity and SMN protein level in spinal muscular atrophy. Nat Genet 1997;16:265–269. 28. Coovert DD, Le TT, McAndrew PE, et al.: The survival motor neuron protein in spinal muscular atrophy. Hum Mol Genet 1997;6:1205–1214. 29. Tiziano FD, Lomastro R, Di Pietro L, et al.: Clinical and molecular cross-sectional study of a cohort of adult type III spinal muscular atrophy patients: clues from a biomarker study. Eur J Hum Genet 2013;21:630–636.
Address correspondence to: Dr. Hannes Steinkellner Institute of Medical Genetics Medical University of Vienna Waehringer Strasse 10 1090 Vienna Austria E-mail: [email protected] Abbreviations Used Ctrl ¼ control CV ¼ coefficient of variation DRG ¼ dorsal root ganglion ECLIA ¼ electrochemiluminescence immunoassay ELISA ¼ enzyme-linked immunosorbent assay FBS ¼ fetal bovine serum hSCs ¼ human Schwann cells MLPA ¼ multiplex ligation-dependent probe amplification MSD ¼ Meso Scale Discovery PBMC ¼ peripheral blood mononuclear cells PBS ¼ phosphate-buffered saline RT ¼ room temperature SD ¼ standard deviation SEM ¼ standard error mean SMA ¼ spinal muscular atrophy SMN ¼ survival motor neuron
ª MARY ANN LIEBERT, INC. VOL. 13 NO. 3 APRIL 2015
ASSAY and Drug Development Technologies 173
Detection of Survival Motor Neuron Protein in Buccal Cells Through Electrochemiluminescence-Based Assay Hannes Steinkellner,1,* Julia Etzler,1,* Bernhard M. Gmeiner,2 and Franco Laccone1 Institutes of 1Medical Genetics and 2Medical Chemistry, Medical University of Vienna, Vienna, Austria. *These two authors contributed equally to this work.
ABSTRACT Spinal muscular atrophy (SMA) is a severe autosomal recessive disorder affecting one in every 10,000 live births. The disease is characterized by loss of alpha-motor neurons in the spinal cord that leads to progressive atrophy and weakness of limb and trunk muscles. This neuromuscular disorder results from deletions and/ or mutations within the survival motor neuron 1 (SMN1) gene, leading to a pathologically decreased expression of functional full-length SMN protein. Here we report on the investigation to measure SMN protein levels through electrochemiluminescence immunoassay (ECLIA). This simple assay is a highly quantitative method able to measure SMN protein levels in human, mouse, and rat samples throughout a wide working range with low intra- and interassay error. The sensitivity for human SMN is 30 pg/mL and provides a new tool for the set up of highthroughput screening for basic research. Moreover, we describe a novel tool for a noninvasive assessment of SMN in buccal cells derived from healthy donors, SMA carriers, and SMA patients. The availability of a validated quantitative ECLIA should improve the investigation of novel compounds for the treatment of SMA.
INTRODUCTION
S
pinal muscular atrophy (SMA) is an autosomal recessive neuromuscular disorder characterized by hypotonia, progressive proximal muscle weakness, and respiratory distress that result from sequential degeneration of alpha-motor neurons in the anterior horn of the spinal cord.1 Although carrier rates vary considerably among different ethnic groups,2 SMA is one of the most common monogenetic disorders, with an incidence of 1 in 6,000 to 1 in 10,000 live births,3,4 and a leading genetic cause
DOI: 10.1089/adt.2015.635
of infant mortality. Based upon the severity of its clinical manifestations, namely the age of disease onset and the motor milestones achieved, the spectrum of SMA can be subdivided into four types.5 In the most severely affected patients (Type I), symptoms manifest by 6 months of age and usually lead to death by respiratory failure before age 2, whereas Type II and III patients present with first symptoms after the age of 6 months and develop the ability to sit (Type II) or walk (Type III), respectively. Comparatively, mildly affected patients with a disease onset after the age of 30 and a normal life expectancy are classified as Type IV SMA. In 1995, the survival motor neuron 1 (SMN1) gene was identified as an SMA-determining gene,1 and it has since been shown that homozygous deletion or mutational inactivation of SMN1 is causative of all four types of SMA. The broad clinical spectrum of the disorder can, in part, be explained by its unique genetic background: the SMN1 gene is located within a duplicated and inverted 500 kb chromosomal segment on 5q13 that comprises a neighboring nearly identical gene called SMN2. The two genes differ in five base pairs only, all of which are located toward their 30 ends,1,6 and they are both ubiquitously expressed. Full-length transcripts derived from both SMN1 and SMN2 encode an identical 294 AA product, the SMN protein. However, due to a critical C-to-T transition at position + 6 of exon 7 of the SMN2 gene, only about 10% of its transcripts are correctly spliced full-length mRNAs, while the vast majority of transcripts derived from SMN2 lack exon 7,5,7,8 resulting in the translation of a truncated SMND7 protein that is highly unstable and cannot functionally replace SMN.9,10 Although the homozygous inactivation of SMN1 in SMA patients cannot be fully compensated for, a small amount of functional SMN protein is still produced by the SMN2 gene. Consequentially, an increased number of SMN2 copies is thought to moderate the lack of SMN protein, thus reducing SMA severity, which explains why SMN2 copy number has been shown to inversely correlate with the clinical severity of the disorder.11–13 In SMA patients, the SMN2 copy number can reach up to eight and usually varies between 1 and 4
ª MARY ANN LIEBERT, INC. VOL. 13
NO. 3 APRIL 2015 ASSAY and Drug Development Technologies
167
STEINKELLNER ET AL.
copies,12 whereas a complete loss of SMN2 in an SMN1 null genome is embryonically lethal in humans14—in other words, every SMA patient possesses at least one copy of SMN2 and is thus expected to produce a basic level of SMN protein. Given the particular genetics that gives rise to SMA, enhancing the production of full-length SMN protein from the SMN2 gene, which is present in all patients, has become a major therapeutic approach.15,16 For any drug discovery program based on that mode of action, a key prerequisite is the reliable and accurate quantification of SMN protein levels in various fluids, cells, and tissues. Moreover, as some of the preclinical drug candidates eventually enter clinical testing, these quantification methods also need to perform rapidly and on a high-throughput basis. To meet these criteria, different SMN-specific enzyme-linked immunosorbent assays (ELISAs) have been developed during the past years that allow for both SMA drug screening and SMN measurement in human and mouse tissue samples, including peripheral blood mononuclear cells (PBMCs).16–19 Although ELISAs can detect much lower protein quantities than classical western blotting (dynamic range of 50– 3,200 pg/mL),19 in terms of sensitivity they are outperformed by a recently developed detection method called electrochemiluminescence immunoassay (ECLIA) whose principle of function is described elsewhere.20,21 The aim of this study was the development and verification of this high-throughput assay for simple and accurate quantification of SMN protein levels that allows a highly sensitive detection limited to 30 pg/mL.
MATERIALS AND METHODS All the samples used in this study were taken after the participants provided their consent by signing a statement of agreement. The SMA carrier provided written informed consent before providing buccal cells used in this study. For the SMA patient, we obtained written informed consent from his guardian.
with embryonic rat dorsal root ganglion neurons),22 human Schwann cells (hSCs, kindly provided by Prof. Ahmet Ho¨ke, John Hopkins University),23 and an SMA fibroblast cell line (#GM22592; Corriel Cell Repositories). The cells were maintained in Dulbecco’s modified Eagle’s medium, including high glucose and stabilized glutamine (Invitrogen), supplemented with 10% fetal bovine serum (FBS, from PAA Laboratories; for hSCs, we used FBS from Sigma-Aldrich) and penicillin/ streptomycin (Gibco/Life Technologies). The hSC medium was further enriched with 0.2% glucose (Gibco) and 2 mM forskolin (Sigma-Aldrich). Cells were incubated in a humidified atmosphere (5% CO2, 37C) for 48 h before sample preparation. Sample Preparation Cells obtained from the culture of different sources were harvested and washed twice with phosphate-buffered saline (PBS, pH 7.4), centrifuged at 250 g for 10 min, and stored at - 80C for further use. In addition, human lymphocytes were obtained from isolated fresh blood samples with LymphoprepTM (Axis-Shield) according to the manufacturer’s procedure. The cell pellets were dissolved in ice-cold lysis reagent (1 · tris-buffered saline pH 7.4, 1% triton X-100, 1 mM ethylenediaminetetraacetic acid, 1 mM ethylene glycol tetraacetic acid, 0.05% sodium dodecyl sulfate, 0.1% sodium deoxycholate, and 2% nonidet-P40) with freshly added protease inhibitors (1 mM phenylmethylsulfonyl fluoride [from Sigma], 5 mM sodium fluoride [Merck Chemicals], 1 mM sodium orthovanadate, and protease inhibitor cocktail [#PI8340 from Sigma Aldrich]) and incubated on ice for about 25 min before centrifugation at 10,000 g for 12 min at 4C. Then, the supernatant was analyzed by the BCA Protein Assay Kit (Thermo Scientific Pierce) for protein concentration.
Antibodies We used a primary mouse SMN monoclonal antibody as the capture antibody from Sigma-Aldrich (clone 2B1, #S2944) and a secondary SMN rabbit polyclonal antibody from Proteintech (#11708-1-AP); the goat anti-rabbit HRP-Sulfo TAGTM antibody from Meso Scale Discovery (MSD) was applied as a detection antibody. The recombinant human SMN1 standard protein was purchased from Enzo Life Sciences (#ADI-NBP-201-050).
Buccal Cell Preparation Buccal cells were harvested using MasterAmp Buccal Swab Brushes (Epicentre Biotechnology) by firmly pressing a brush against the inner cheek and swabbed while twirling for 60 s on each cheek. Both brushes were then immersed in the same tube containing 500 mL of ice-cold PBS and twirled gently to dislodge the cells. Samples were then microcentrifuged at 10,000 g for 10 min at 4C, and the cell pellet was lysed in the ice-cold lysis reagent as described in the sample preparation section. The protein concentration was calculated using the BCA protein assay as described above.
Cell Culture The cell lines used for the experiments were F11 (a fusion product of the mouse neuroblastoma cell line N18TG-2
SMN ECLIA Protocol The capture antibody was coated in the 96-well Multi-Array High Bind plates (MSD) by overnight incubation at 4C. Next
168 ASSAY and Drug Development Technologies APRIL 2015
ª MARY ANN LIEBERT, INC.
SMN PROTEIN AND ECLIA ASSAY
day, the plates were blocked at room temperature with 125 mL MSD Blocker A. The plates were washed thrice with PBSTween (0.05%) before adding the samples (25 mL per well diluted in 1% MSD Blocker or cell lysis buffer). The sample preparations were incubated at room temperature for another 90 min or at 4C for 16 h in case of buccal cell lysates. Each well was incubated with the polyclonal rabbit anti-SMN antibody (0.5 mg/mL in 1% MSD Blocker A) for 75 min followed by one incubation of 0.75 mg/mL of MSD Sulfo-TAG goat anti-rabbit detection antibody in 1% MSD Blocker A. One hundred fifty microliter Tris-based Read Buffer T (1 · ) with surfactant (MSD) was added to the plate and signals from ECL were captured by the Sector Imager 2400 reader (MSD) and recorded as signal counts (Table 1). SMN Copy Number Quantification of SMN1 and SMN2 copy number in four healthy donors was performed by Multiplex Ligation-
Table 1. Spinal Muscular Atrophy Electrochemiluminescence Immunoassay Protocol Table Step
Parameter
Value
Description
Fig. 1. Standard curve of recombinant human SMN1. Error bars show standard error at each dilution (n = 3). Linear regression in a range from 0.76 to 5,000 pg per well was generated using GraphPad Prism software. SMN, survival motor neuron.
dependent Probe Amplification (MLPA) using P021-A2 and P060-B2 SALSA MLPA probemix (MRC-Holland) according to the manufacturer’s instructions. Genomic DNA was extracted from blood lymphocytes by standard salt precipitation,24 and a total of 150 ng was used in each reaction. MLPA was run in duplicate for each donor and four different DNA standards.
1
Capture antibody
25 mL
1:4,000 in PBS buffer
2
Incubation
Overnight
Seal the plate, 4C
3
Blocking
125 mL
MSD Blocker A in PBS buffer
4
Incubation
90 min
Seal the plate, RT, shaking
5
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
6
Standard/samples
25 mL
In MSD Blocker A or Lysis buffer
7
Incubation
90 min
Seal the plate, RT, shaking
8
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
Table 2. Working Ranges of Different Sample Materials
9
Detection antibody
25 mL
1:3,330 in PBS buffer
Sample type
10
Incubation
75 min
Seal the plate, RT, shaking
HepG2, HEK 293
11
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
Human fibroblasts: SMA patients/control
1–5
12
Sulfo-TAG
25 mL
1:500 in 1% MSD Blocker A in PBS buffer
Murine fibroblasts
1–5
Buccal cells: patients/control
1–10
13
Incubation
60 min
Seal the plate, RT, shaking
14
3· washing
150 mL
0.05% TWEEN 20 in PBS buffer
15
Read Buffer T
150 mL
1:4 in Aqua Dest
16
Assay readout
CCD camera
MSD Sector Imager 2400
Statistical Analysis Statistical analysis of the measurement data was performed with GraphPad Prism software. Results are presented as mean – standard deviation and mean – standard error of the means. Statistical significance of differences was examined by using
PBS, phosphate-buffered saline; RT, room temperature; MSD, Meso Scale Discovery.
ª MARY ANN LIEBERT, INC. VOL. 13
Working range [lg] 0.1–2
F11 Rat DRG cells
0.5–2
Human Schwann cells
0.1–2
Mouse brain lysates
1–4
Suggested working ranges for various human, mouse, and rat sample materials. SMA, spinal muscular atrophy; DRG, dorsal root ganglion.
NO. 3 APRIL 2015
ASSAY and Drug Development Technologies 169
STEINKELLNER ET AL.
Table 3. Intra-Assay and Interassay Precision of Survival Motor Neuron Electrochemiluminescence Immunoassay pg SMN/lg Protein Well 1
Well 2
Well 3
Mean
SD
SEM
%CV
F11—rat DRG cells
1390.14
1425.14
1405.29
1406.86
17.55
10.13
1.25
Human Schwann cells
2223.10
2209.8l
2223.10
2245.35
38.54
22.25
1.72
PBMCs
8.57.32
891.78
846.49
865.20
23.65
13.65
2.73
Intra-assay
in the reactivity of the assay toward the SMN protein of different species. The measurement of different samples on three ensuing days yields an intraday precision of coefficient of variation (CV) £ 2.73% and an interday precision of CV £ 7.09% (Table 3).
Quantification of SMN Protein Levels in Human Buccal Cells To investigate whether our ECLIA would SMA fibroblasts 855.51 859.30 841.62 852.15 9.31 5.38 1.09 be sensitive enough to quantify the SMN Interassay protein in human buccal cells, we isolated buccal cells from one male healthy F11—rat DRG cells 1367.09 55.94 32.30 4.09 adult. The protocol adapted to this was as Human Schwann cells 2161.07 121 99 70.43 5.54 described in the Materials and Methods section because higher accuracy was PBMCs 702.54 23.85 16.86 3.39 achieved by overnight incubation of the SMA fibroblasts 835.93 59.24 34.20 7.09 samples. Intra-assay precision (n = 3 wells per day) and interassay precision (n = 3 separate days). The linearity over a wide range of SD, standard deviation; SEM, standard error mean; CV, coefficient of variation. buccal cell lysates from a healthy donor prompted us to investigate SMN protein levels in a small cohort of healthy donors (n = 4, Fig. 2). For the Student’s t-test. Significant differences are marked in the this, we collected buccal cells from an unmatched group of figures with *(P < 0.05), **(P < 0.01), and ***(P < 0.001). controls thrice within a 48-h interval (Table 4). Although we found a high variability in SMN protein levels RESULTS (Fig. 3) in this group of healthy donors, we investigated the Assay Specificity, Reproducibility, and Sensitivity possibility to measure SMN protein levels of buccal cell lysates The ECLIA is based on 96-microwell plates with carbon in an SMA carrier and patient. We collected buccal cells from electrodes built into the bottom for ultrasensitive detec25 one SMA carrier (male, 36 years old) and one SMA Type III tion. A standard curve with recombinant human SMN1 is shown in Figure 1. Recombinant human SMN1 can be measured over a wide range (0.76–5,000 pg per well with R2 = 0.9983) with high accuracy, which equals a dynamic range of 30–200,000 pg/mL. The recovery rate was estimated by spiking a definite amount of different cell lysates with low and high amounts of recombinant human SMN1. The recovery rate of spiked recombinant SMN1 with human SMA fibroblasts was 80% (high amounts) and 90% (low amounts). Similarly, we estimated a recovery rate for F11 cells of 79% (high amounts) and 93% (low amounts) (Data not shown). Validation of the Assay The SMN ECLIA is able to quantify SMN protein levels in a variety of cell lines and tissues of human, mouse, and rat origin. The working range was found to vary among different sample types (Table 2), which on one hand is most likely proportional to the concentrations of SMN per mg protein in different tissues and on the other hand also reflects differences
170 ASSAY and Drug Development Technologies APRIL 2015
Fig. 2. Linearity of buccal cell lysate. The assay remains sensitive and in the linear range (R2 = 0.98) from 0.3 to 4 mL buccal cell lysate per well.
ª MARY ANN LIEBERT, INC.
SMN PROTEIN AND ECLIA ASSAY
Table 4. Demographic Values of Healthy Donors, One Spinal Muscular Atrophy Carrier, and One Spinal Muscular Atrophy Patient Control (No.)
Sex
Age
Copies (SMN1/SMN2)
1
Female
29
2/2
2
Male
31
2/2
3
Male
61
2/2
4
Male
55
2/3
SMA carrier
Male
35
1/3
SMA patient
Male
6
-/3
Individual values of age (years) at the study, sex, and copy numbers of SMN1 and SMN2 are shown. No genotyping was performed on SMA carriers and SMA patients in this study, and the fathers self-identified themselves as SMA carriers and patients based on prior testing and diagnoses. SMA, spinal muscular atrophy.
patient (male, 6 years old) and measured SMN protein levels from buccal cell lysates with the ECLIA-based assay. As shown in Figure 4, SMN was readily detected in buccal cell lysates from the SMA carrier and a Type III SMA patient. The SMN protein level in the SMA carrier was slightly reduced compared to a sex- and age-matched healthy donor (Ctrl 2). The amount of SMN protein in the SMA patient was highly increased in relation to both, the control and the carrier, but could not be compared to an age-matched healthy control in this preliminary assessment.
Fig. 3. SMN protein levels in human buccal cells from healthy adults. Buccal cells were collected preprandial thrice from four healthy adults (Ctrl 1–4) on five consecutive days at every 48 h. SMN protein levels were measured in buccal cell extracts and normalized to protein levels. The data presented are mean – SD of triplicate wells (n = 3). Ctrl, control; SD, standard deviation.
Fig. 4. SMN protein levels in human buccal cells from SMA carrier and SMA patient. SMN protein levels in human buccal cells derived from SMA carrier and SMA patient compared to healthy control (Ctrl 2). The data presented are mean – SD of triplicate wells (n = 3). SMA, spinal muscular atrophy.
DISCUSSION We developed and validated an ECLIA for the quantitative measurement of SMN protein levels in human, mouse, and rat samples. Human SMN, which is ubiquitously expressed, has a molecular weight of 38 kDa and localizes to both the cytoplasm and the nucleus. It is part of a multiprotein complex that contains multiple SMN molecules and a number of stoichiometrically interacting proteins, Gemins 2–8.12 As with other SMN quantification methods, lysis reagents used in the ECLIA protocol have to be chosen carefully to ensure extraction and dissociation of both cytoplasmic and nuclear SMN.16 In the protocol described here, a commercially available lysis buffer known to work well in the ECLIA setting (MSD) is used, slightly modified for SMN extraction. Compared to other extraction buffers used in preliminary trials, it yielded higher concentrations of SMN in relation to whole protein (data not shown), suggesting a more efficient extraction of nuclear SMN. Both antibodies adapted for our ECLIA protocol are used in a well-established ELISA that has recently been used for measurement of blood SMN levels in a large clinical SMA study.13,19 The primary antibody (2B1) is a mouse monoclonal antibody that binds to the N-terminus of human SMN at amino acid positions 14–2019,26; the epitope for the secondary antibody (11708-1-AP), which is a rabbit polyclonal antibody, is located between amino acids 197 and 204 of human SMN.19 As both epitopes are expected to be present in full-length SMN as well as in SMND7, it is conceivable that both proteins are detected by our ECLIA, thus potentially interfering with the measurement of slight shifts in the ratio of SMN/SMND7.
ª MARY ANN LIEBERT, INC. VOL. 13 NO. 3 APRIL 2015
ASSAY and Drug Development Technologies 171
STEINKELLNER ET AL.
This should be kept in mind when using the ECLIA in preclinical screenings for drugs that increase the rate of exon7inclusion in SMN2-derived transcripts. Within the context of clinical studies, however, SMND7 is unlikely to substantially distort the quantification of SMN in tissue samples, as the truncated protein contains a potent degradation signal that causes it to be rapidly degraded.10 With regard to the assessment of SMN levels in murine samples using ELISA, concerns have been raised by Kobayashi et al. as to the differences in sensitivity. The authors report on a marked difference in reactivity of the assay to the SMN protein in humans and mice, respectively, with recombinant mouse SMN being detected at only about 10% the levels of recombinant human SMN.19 It is reasonable to assume that this holds true for the ECLIA as well, as the two assays are based on the exact same primary and secondary antibody. The ECLIA described here has a broad dynamic range (0.76– 5,000 pg/well with R2 = 0.9983) and a detection limit of 30 pg/ mL for the human SMN protein. It is therefore more sensitive than sandwich format ELISAs, whose reported detection limits range from 25 ng/mL18 to 50 pg/mL.19 As a result, SMN levels of various origin can be reliably assessed even at great dilutions, which means the ECLIA is less prone to matrix effects than other protein quantification methods and thus less subject to intersample variation. In line with its enhanced sensitivity, the ECLIA readily detected SMN in buccal swabs of a Type III SMA patient, an adult SMA carrier, and a healthy adult donor. To our knowledge, this is the first report of SMN protein quantification in human buccal cells. As they are particularly easy to access, it is tempting to speculate that buccal cells might constitute a noninvasive and practicable way to monitor SMN levels in future trials of SMNupregulating therapeutics. However, earlier studies using ELISA to assess SMN levels in different human tissue types have pointed at various pitfalls with regard to that approach. First, quantitative evaluation of murine tissues suggest that Smn expression varies widely across tissues of the same genotype, with Smn levels being found to be lowest in the primary tissues associated with SMA, that is, nerve, spinal cord, and muscle.19 It has been pointed out that SMN levels measured in PBMCs do not necessarily fully reflect SMN expression in more disease-relevant cells like motor neurons or muscle cells.13 The same holds true for buccal cells and any other peripheral tissues. Second, age is an important factor to consider when assessing SMN levels in clinical specimens, as there is evidence of a general decline of SMN levels across tissue types and species,16 and several studies using ECLIA to measure SMN levels in PBMCs of SMA patients, carriers, and controls have
172 ASSAY and Drug Development Technologies APRIL 2015
stressed the necessity to control for age when performing any statistical analysis.13,19 Third, even during a short period of time, SMN levels measured in PBMCs fluctuate substantially within individuals; when assessed at five time points over the course of 1 month, blood SMN levels were shown to vary from 2 to 10 times in six different healthy adults.19 Whether SMN levels measured in buccal cells using ECLIA are subject to such an extent of intraindividual fluctuation has to be evaluated by repeated within-subject testing, preferably normalizing SMN signals by cell counts instead of total protein. Finally, the finding that there is no clear correspondence between SMN protein levels in blood cells and disease severity raises fundamental concerns with regard to using peripheral SMN protein levels as an indicator of response to SMA treatment; while a correlation of protein quantity with phenotype could be established for severely affected Type I and Type II patients, no substantial decrease in the SMN protein is detectable in Type III patients.27,28 In line with this observation, a recent multicenter clinical trial using SMN ELISA to assess SMN levels in PBMCs of adult Type III SMA patients did not show any correlation between protein quantity and motor performance, leading the authors to suggest that SMN protein levels in blood cells are not suitable as biomarkers in clinical trials on SMA.29 In light of the data acquired from Type III SMA blood cells, our preliminary finding of a high SMN protein level in buccal cells of one Type III SMA patient is not that surprising, given the age difference between the patient (6 years) and the healthy control (31 years). It is doubtful that SMN levels in buccal cells will provide a more reliable indicator of disease severity in a mildly affected SMA patient than PBMCs, thus compromising the use as a biomarker in SMN upregulation trials. Taken together, the use of ECLIA to monitor SMN levels in buccal cells as a biomarker in clinical SMA drug trials is far from being straightforward. However, our data suggest that it can be effectively used to measure SMN protein levels as well as slight changes thereof in cell models of different species with high accuracy and specificity over a wide range. In terms of sensitivity, the ECLIA outperforms previous quantification methods. Furthermore, due to its relatively easy handling in a 96-well plate format, the assay may represent a novel tool of great value for high-throughput drug discovery screenings.
ACKNOWLEDGMENTS We gratefully thank Dr. Mark Du¨rr from Meso Scale Discovery (MSD, Germany) for his help with the ECLIA optimization and Dr. Oliver Brandau for collection of blood samples. Furthermore, we thank all donors for buccal cell samples.
ª MARY ANN LIEBERT, INC.
SMN PROTEIN AND ECLIA ASSAY
AUTHOR CONTRIBUTIONS All the authors of this article fulfilled the criteria of authorship. H.S. performed the experiments. H.S., J.E., B.M.G., and F.L. designed the study and wrote the article.
DISCLOSURE STATEMENT No competing financial interests exist.
REFERENCES 1. Lefebvre S, Burglen L, Reboullet S, et al.: Identification and characterization of a spinal muscular atrophy-determining gene. Cell 1995;80:155–165. 2. Hendrickson BC, Donohoe C, Akmaev VR, et al.: Differences in SMN1 allele frequencies among ethnic groups within North America. J Med Genet 2009; 46:641–644. 3. Pearn J: Incidence, prevalence, and gene frequency studies of chronic childhood spinal muscular atrophy. J Med Genet 1978;15:409–413. 4. Crawford TO, Pardo CA: The neurobiology of childhood spinal muscular atrophy. Neurobiol Dis 1996;3:97–110. 5. Wirth B, Brichta L, Hahnen E: Spinal muscular atrophy: from gene to therapy. Semin Pediatr Neurol 2006;13:121–131. 6. Burglen L, Lefebvre S, Clermont O, et al.: Structure and organization of the human survival motor neurone (SMN) gene. Genomics 1996;32:479–482. 7. Monani UR, Lorson CL, Parsons DW, et al.: A single nucleotide difference that alters splicing patterns distinguishes the SMA gene SMN1 from the copy gene SMN2. Hum Mol Genet 1999;8:1177–1183. 8. Lorson CL, Hahnen E, Androphy EJ, Wirth B: A single nucleotide in the SMN gene regulates splicing and is responsible for spinal muscular atrophy. Proc Natl Acad Sci U S A 1999;96:6307–6311. 9. Lorson CL, Androphy EJ: An exonic enhancer is required for inclusion of an essential exon in the SMA-determining gene SMN. Hum Mol Genet 2000;9: 259–265. 10. Cho S, Dreyfuss G: A degron created by SMN2 exon 7 skipping is a principal contributor to spinal muscular atrophy severity. Genes Dev 2010;24:438–442. 11. Feldkotter M, Schwarzer V, Wirth R, Wienker TF, Wirth B: Quantitative analyses of SMN1 and SMN2 based on real-time lightCycler PCR: fast and highly reliable carrier testing and prediction of severity of spinal muscular atrophy. Am J Hum Genet 2002;70:358–368. 12. Wirth B, Brichta L, Schrank B, et al.: Mildly affected patients with spinal muscular atrophy are partially protected by an increased SMN2 copy number. Hum Genet 2006;119:422–428. 13. Crawford TO, Paushkin SV, Kobayashi DT, et al.: Evaluation of SMN protein, transcript, and copy number in the biomarkers for spinal muscular atrophy (BforSMA) clinical study. PLoS One 2012;7:e33572. 14. Passini MA, Cheng SH: Prospects for the gene therapy of spinal muscular atrophy. Trends Mol Med 2011;17:259–265. 15. Lorson CL, Rindt H, Shababi M: Spinal muscular atrophy: mechanisms and therapeutic strategies. Hum Mol Genet 2010;19:R111–R118. 16. Cherry JJ, Kobayashi DT, Lynes MM, et al.: Assays for the identification and prioritization of drug candidates for spinal muscular atrophy. Assay Drug Dev Technol 2014;12:315–341. 17. Nguyen thi M, Humphrey E, Lam LT, et al.: A two-site ELISA can quantify upregulation of SMN protein by drugs for spinal muscular atrophy. Neurology 2008;71:1757–1763. 18. Piepers S, Cobben JM, Sodaar P, et al.: Quantification of SMN protein in leucocytes from spinal muscular atrophy patients: effects of treatment with valproic acid. J Neurol Neurosurg Psychiatry 2011;82:850–852.
19. Kobayashi DT, Olson RJ, Sly L, et al.: Utility of survival motor neuron ELISA for spinal muscular atrophy clinical and preclinical analyses. PLoS One 2011;6: e24269. 20. Yazuburo Namba MU, Osamu Suzuki: Highly sensitive electrochemiluminescence immuneassay using the ruthenium chelate-labeled antibody bound on the magnetic microbeads. Anal Sci 1999;15:1087–1093. 21. Debad JB, Glezer EN, Leland JK, Wohlstadter J, Sigal GB: Clinical and biological applications of ECL. In: Electrogenerated Chemiluminescence. Bard AJ (ed.), pp. 359–396. Marcel Dekker, New York, 2004. 22. Fan SF, Shen KF, Scheideler MA, Crain SM: F11 neuroblastoma · DRG neuron hybrid cells express inhibitory mu- and delta-opioid receptors which increase voltage-dependent K + currents upon activation. Brain Res 1992;590:329–333. 23. Lehmann HC, Chen W, Mi R, et al.: Human Schwann cells retain essential phenotype characteristics after immortalization. Stem Cells Dev 2012;21:423– 431. 24. Miller SA, Dykes DD, Polesky HF: A simple salting out procedure for extracting DNA from human nucleated cells. Nucleic Acids Res 1988;16:1215. 25. Steinkellner H, Scheiber-Mojdehkar B, Goldenberg H, Sturm B: A high throughput electrochemiluminescence assay for the quantification of frataxin protein levels. Anal Chim Acta 2010;659:129–132. 26. Liu Q, Dreyfuss G: A novel nuclear structure containing the survival of motor neurons protein. EMBO J 1996;15:3555–3565. 27. Lefebvre S, Burlet P, Liu Q, et al.: Correlation between severity and SMN protein level in spinal muscular atrophy. Nat Genet 1997;16:265–269. 28. Coovert DD, Le TT, McAndrew PE, et al.: The survival motor neuron protein in spinal muscular atrophy. Hum Mol Genet 1997;6:1205–1214. 29. Tiziano FD, Lomastro R, Di Pietro L, et al.: Clinical and molecular cross-sectional study of a cohort of adult type III spinal muscular atrophy patients: clues from a biomarker study. Eur J Hum Genet 2013;21:630–636.
Address correspondence to: Dr. Hannes Steinkellner Institute of Medical Genetics Medical University of Vienna Waehringer Strasse 10 1090 Vienna Austria E-mail: [email protected] Abbreviations Used Ctrl ¼ control CV ¼ coefficient of variation DRG ¼ dorsal root ganglion ECLIA ¼ electrochemiluminescence immunoassay ELISA ¼ enzyme-linked immunosorbent assay FBS ¼ fetal bovine serum hSCs ¼ human Schwann cells MLPA ¼ multiplex ligation-dependent probe amplification MSD ¼ Meso Scale Discovery PBMC ¼ peripheral blood mononuclear cells PBS ¼ phosphate-buffered saline RT ¼ room temperature SD ¼ standard deviation SEM ¼ standard error mean SMA ¼ spinal muscular atrophy SMN ¼ survival motor neuron
ª MARY ANN LIEBERT, INC. VOL. 13 NO. 3 APRIL 2015
ASSAY and Drug Development Technologies 173
