Summary
A new procedure was developed and applied to study immunoglobulin free light chains (FLC) in saliva of healthy subjects and patients with multiple sclerosis (MS). The procedure was based on a Western blot analysis for detection and semiquantitative evaluation of monomeric and dimeric FLCs. The FLC indices accounting for the total FLC levels and for the monomer/dimer ratios of κ and λ FLC were calculated, and the cut‐off values of the FLC indices were determined to distinguish healthy state from MS disease. The obtained FLC index values were statistically different in the saliva of three groups: active MS patients, MS patients in remission and healthy subjects groups. Our FLC monomer–dimer analysis allowed differentiation between healthy state and active MS with specificity of 100% and a sensitivity of 88·5%. The developed technique may serve as a new non‐invasive complementary tool to evaluate the disease state by differentiating active MS from remission with sensitivity of 89% and specificity of 80%.
Keywords: dimers, immunoglobulin free light chains, monomers, multiple sclerosis, saliva
Introduction
Multiple sclerosis (MS) is a putatively autoimmune inflammatory disease of the central nervous system (CNS) affecting mainly relatively young adults. The diagnosis of MS is based on clinical and paraclinical tests including magnetic resonance imaging, as well as analysis of cerebrospinal fluid (CSF), where demonstration of oligoclonal immunoglobulins (Ig) indicates intrathecal Ig production typical of MS. Although the oligoclonality test is used commonly in MS management, the oligoclonal Ig might be detected in other inflammatory and infectious CNS diseases, thus reducing the utility of this method for differential diagnosis.
During the last decade, the intrathecal production of not only intact Ig, but also of Ig free light chains (FLC), gained considerable interest in the diagnosis of MS. An increasing body of evidence suggests strongly that intrathecal production of FLC is markedly elevated in MS, and that quantification of FLC in CSF may contribute to the diagnosis 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 and to the understanding of the pathogenesis of MS 11, 12.
Recently, we have developed a new Western blot‐based technique to analyse FLC monomers and dimers in CSF of MS patients 3, 4, 5, 6, 7. Both specificity and sensitivity of our diagnostic method of MS are higher than those of the conventionally used oligoclonality test 3, 7. In addition, we observed three abnormal FLC patterns in MS patients, characterized by increased levels of either κ FLC (monomers and dimers) or λ FLC dimers, or both (i.e. κ monomers and dimers plus λ dimers). The clinical significance of such differences in FLC patterns is currently under investigation.
Although useful in diagnosis and management, the repeated testing of CSF for the follow‐up of MS patients is impractical for being invasive. We hypothesized that FLC analysis of the patients' saliva might be more useful for this purpose. This assumption is based on earlier reports on changes in mucosal immunity in MS, including: increased numbers of immunoreactive cells in saliva and tears 13, 14, impairment of mucosal barrier of the gut (‘leaky gut’) 15 and presence of oligoclonal Ig bands and increased amounts of monomeric IgA and IgG in tears 16, 17, 18.
In the current study, we examine the efficacy of the saliva FLC testing for the diagnosis and disease activity evaluation in MS. We report a new technique for analysis of the monomeric and dimeric FLC in saliva and compare the FLC monomer‐dimer patterns in MS patients with those of healthy subjects.
Materials and methods
Study overview
Here we describe the development of a non‐invasive practical test, based on a semiquantitative Western blot analysis of FLCs in saliva. The developed technique was used to evaluate total levels of saliva FLCs and the ratios of FLC monomer and dimers. We found this technique to be useful in distinguishing between patients with active MS and healthy subjects, and between MS patients with active disease and those in remission. The study was approved by the institutional review board.
Patients and samples
Patients with a definite diagnosis of MS (n = 85, aged 20–65 years, mean age 45·3 ± 13·2) were studied. MS diagnosis was based on revised McDonald criteria 19. Seventy‐three of 85 MS patients showed the relapsing–remitting form of the disease (RR‐MS), and the remaining 12 patients had a secondary progressive course (SP‐MS). MS patients were under at least one of the following treatments: interferon beta‐1a (IFN‐β1a), interferon beta‐1b, dimethyl fumarate, fingolimod, glatiramer acetate and naltrexone (Table 1).
Table 1.
Demographic and clinical characteristics of multiple sclerosis (MS) patients
| RR‐MS | SP‐MS | |
|---|---|---|
| Number of patients | 73 | 12 |
| Gender | m = 26, f = 47 | m = 4, f = 8 |
| Age (mean ± s.d.) | 43·2 ± 12·8 | 57·9 ± 6·8 |
| Age of disease onset (mean ± s.d.) | 34·3 ± 12·2 | 37·1 ± 7·3 |
| Disease duration (in years) prior to saliva collection (mean ± s.d.) | 10·3 ± 7·9 | 19·9 ± 10·3 |
| EDSS on the day of saliva collection (mean ± s.d.) | 2·0 ± 1·0 | 6·3 ± 0·9 |
| Disease‐modifying drug (DMD) treatment | ||
| Interferon β‐1b (Betaferon) | 25 | 3 |
| Interferon β‐1a (Rebif, Avonex) | 29 | 3 |
| Glatiramer acetate (Copaxon) | 4 | 1 |
| Dimethyl fumarate (Tecfidera) | 5 | |
| Fingolimod (Gilenya) | 4 | |
| Low‐dose naltrexone | 1 | |
RR‐MS = relapsing–remitting MS; SP‐MS = secondary progressive MS; m = male; f = female; s.d. = standard deviation; EDSS = expanded disability status scale.
Saliva samples of MS patients were collected during their regular visits at the MS Clinical Research and Treatment Service of a large tertiary Medical Center. On the day of saliva sampling patients underwent clinical evaluation to determine their expanded disability status scale (EDSS) scores. Also, the EDSS scores were determined 3 months prior to and after sampling, as part of patients' evaluation during their regular follow‐up visits. Based on patients' files, three subgroups of MS patients were defined: the RR‐MS patients in remission (n = 58), the RR‐MS patients with relapse of the disease (n = 15) and SP‐MS patients (n = 12) (Tables 2, 3, 4). Patients with relapse showed worsening of old symptoms or the appearance of new MS symptoms that lasted more than 24 h. Some patients with relapse received a course of treatment with steroids, as indicated in Table 2.
Table 2.
Patients with relapsing–remitting multiple sclerosis (RR‐MS: clinical data and free light chain (FLC) analysis
| EDSS | Saliva FLC analysis | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Patient no. | Gender | Age, years | Disease duration, years | Relapse duration, days | Application of steroids, start | At sampling | 3 months before sampling | 3 months after sampling | Total FLC level index | κ M/D ratio index | λ M/D ratio index | Supports active MS or remission * |
| 1 | f | 45 | 10 | 60 | Prior sampling | 2·5 | 2 | 2·5 | 18·3 | 3·2 | 3·5 | Active |
| 2 | f | 43 | 23 | 15 | Not applied | 2·5 | 3 | 3 | 42·0 | 15·8 | 13·0 | Active |
| 3 | m | 51 | 0·5 | 45 | Not applied | 3 | 2·5 | 2 | 16·6 | 6·5 | 7·7 | Active |
| 4 | f | 43 | 20 | 35 | Not applied | 2·5 | 3·5 | 2·5 | 24·2 | 2·0 | 0·9 | Active |
| 5 | f | 42 | 20 | 30 | After sampling | 2·5 | 1·5 | 2 | 3·3 | 0·6 | 2·4 | Borderline |
| 6 | m | 39 | 5 | 21 | Prior sampling | 2·5 | 2·5 | 2 | 19·3 | 1·0 | 2·8 | Active |
| 7 | f | 59 | 10 | 30 | Not applied | 6 | 6 | n.a. | 20·1 | 5·5 | 2·4 | Active |
| 8 | m | 25 | 0·3 | 14 | After sampling | 0 | 0 | 0 | 3·9 | 1·9 | 9·6 | Active |
| 9 | f | 59 | 25 | 20 | Not applied | 1·5 | 1 | 2·5 | 73·7 | 1·5 | 14·8 | Active |
| 10 | f | 59 | 18 | 25 | Not applied | 2·5 | 2 | 2·5 | 29·0 | 5·4 | 5·5 | Active |
| 11 | f | 20 | 3 | n.a. | n.a. | n.a. | n.a. | n.a. | 41·5 | 2·6 | 0·8 | Active |
| 12 | f | 58 | 24 | n.a. | n.a. | n.a. | n.a. | n.a. | 19·3 | 5·8 | 25·5 | Active |
| 13 | m | 45 | n.a. | n.a. | n.a. | n.a. | n.a. | n.a. | 45·2 | 26·6 | 22·3 | Active |
| 14 | m | 22 | 2 | 14 | Prior sampling | 1·5 | 2 | 1·5 | 26·1 | 3·5 | 10·6 | Active |
| 15 | f | 35 | 20 | 120 | Not applied | 1 | 1 | 1 | 29·6 | 0·9 | 2·4 | Active |
*Active disease versus remission was determined using the following cut‐off values of the free light chain (FLC) indices: total FLC level = 17, κ M/D ratio = 4·0 and λ M/D ratio = 2·4. Active disease is supported when at least one of these three indices is above the cut‐off value. M = monomer; D = dimer; SP‐MS = secondary progressive MS; EDSS = expanded disability status scale; n.a. = data not available; m = male; f = female.
Table 3.
Patients with secondary progressive multiple sclerosis (SP‐MS): clinical data and free light chain (FLC) analysis
| EDSS | Saliva FLC analysis | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Patient no. | Gender | Age, years | Disease duration, years | At sampling | 3 months before sampling | 3 months after sampling | Total FLC level index | κ M/D ratio index | λ M/D ratio index | Supports active MS or remission * |
| 16 | m | 50 | 22 | 4·5 | 4·5 | 4·5 | 3·5 | 1·0 | 0·7 | Remission |
| 17 | m | 62 | n.a. | 7 | 6·5 | 7 | 36·5 | 10·0 | 4·2 | Active |
| 18 | f | 56 | 20 | 6 | 6·5 | 7 | 19·1 | 1·9 | 3·6 | Active |
| 19 | m | 58 | 17 | 7 | 7 | 7 | 7·5 | 33·4 | 21·9 | Active |
| 20 | f | 65 | 26 | 6 | 6 | 6 | 27·5 | 6·3 | 0·5 | Active |
| 21 | f | 64 | 36 | 7·5 | 7·5 | 7·5 | 58·4 | 1·5 | 0·8 | Active |
| 22 | f | 58 | n.a. | 7 | 7·5 | 7 | 6·1 | 5·4 | 1·7 | Active |
| 23 | f | 60 | 26 | 5·5 | 5 | 5 | 192·1 | 12·6 | 61·2 | Active |
| 24 | f | 59 | 20 | 6 | 6 | 6·5 | 71·7 | 6·4 | 1·4 | Active |
| 25 | m | 41 | 3 | 6 | 5 | 6 | 1·5 | 1·7 | 0·5 | Remission |
| 26 | f | 57 | 3 | 6 | 6 | 6 | 13·1 | 0·6 | 8·6 | Active |
| 27 | f | 65 | 26 | 7·5 | 7·5 | n.a. | 5·0 | 2·9 | 2·3 | Remission |
*Active disease versus remission was determined using the cut‐off values of the following indices: total FLC level = 17, κ M/D ratio = 4·0 and λ M/D ratio = 2·4. Active disease is supported when at least one of these three indices' values is above the cut‐off value. M = monomer; D = dimer; RR‐MS = relapsing–remitting MS; EDSS = expanded disability status scale; n.a. = data not available; m = male; f = female.
Table 4.
Relapsing–remitting multiple sclerosis (RR‐MS) patients in remission: clinical data and free light chain (FLC) analysis
| EDSS | Saliva FLC analysis | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Patient no. | Gender | Age, years | Disease duration, years | At sampling | 3 months before sampling | 3 months after sampling | Total FLC level index | κ M/D ratio index | λ M/D ratio index | Supports active MS or remission * |
| 28 | m | 60 | 3 | 2 | 2 | 2 | 7·2 | 2·6 | 1·1 | Remission |
| 29 | f | 41 | 16 | 1·5 | 2 | 3 | 6·4 | 1·5 | 1·3 | Remission |
| 30 | f | 58 | 5 | n.a. * | n.a. | n.a. | 12·8 | 3·4 | 3·3 | Active |
| 31 | f | 32 | 0·5 | 2 | 2 | 2 | 3·3 | 1·0 | 0·4 | Remission |
| 32 | f | 61 | 8 | 3·5 | 3·5 | 3·5 | 12·1 | 1·6 | 2·5 | Borderline |
| 33 | f | 56 | 19 | 2 | 1·5 | 1·5 | 6·7 | 2·4 | 0·5 | Remission |
| 34 | m | 37 | 12 | 1·5 | 1 | 1·5 | 2·8 | 2·3 | 0·9 | Remission |
| 35 | m | 60 | 9 | 3·5 | 3·5 | 3·5 | 9·4 | 2·3 | 0·5 | Remission |
| 36 | f | 27 | 3 | 1 | 1 | 1 | 2·8 | 2·3 | 0·8 | Remission |
| 37 | m | 22 | 1 | 2 | 2 | 1 | 8·8 | 0·8 | 1·1 | Remission |
| 38 | f | 60 | 21 | 2 | 2 | 2 | 2·3 | 2·3 | 1·1 | Remission |
| 39 | f | 27 | 27 | 2·5 | 2·5 | 2·5 | 44·7 | 10·8 | 0·4 | Active |
| 40 | m | 44 | 18 | 1 | 2 | 0 | 24·2 | 7·6 | 1·0 | Active |
| 41 | m | 41 | 17 | 1·5 | 1 | 1·5 | 0·9 | 4·0 | 1·7 | Borderline |
| 42 | m | 21 | 2 | 1 | 1 | 1 | 14·8 | 1·4 | 2·4 | Borderline |
| 43 | m | 47 | 11 | 2 | 1 | 2 | 8·3 | 0·8 | 1·0 | Remission |
| 44 | m | 49 | 17 | 1·5 | 1 | 2 | 8·0 | 1·0 | 1·2 | Remission |
| 45 | f | 41 | 21 | 2 | 2 | 2 | 37·3 | 0·4 | 8·4 | Active |
| 46 | f | 49 | 21 | 3 | 2·5 | 2·5 | 10·5 | 2·7 | 0·3 | Remission |
| 47 | m | 58 | 20 | 2 | 1·5 | 1·5 | 7·0 | 3·8 | 0·5 | Remission |
| 48 | f | 36 | 2 | 1 | 2 | 1 | 10·0 | 1·5 | 0·7 | Remission |
| 49 | f | 48 | 1 | 2 | 2 | 2 | 4·0 | 1·2 | 0·5 | Remission |
| 50 | f | 22 | n.a. | 1 | n.a. | n.a. | 6·9 | 3·1 | 1·5 | Remission |
| 51 | m | 46 | 15 | 2 | 2 | 2 | 9·9 | 1·8 | 0·4 | Remission |
| 52 | f | 29 | 2 | 1·5 | 1 | 1 | 26·5 | 10·3 | 0·6 | Active |
| 53 | f | 32 | n.a. | 3·5 | 4 | 4 | 7·5 | 0·3 | 1·1 | Remission |
| 54 | f | 34 | 3 | 1·5 | 4 | 1·5 | 6·5 | 0·9 | 1·1 | Remission |
| 55 | f | 37 | 5 | 2 | 1 | 1 | 7·3 | 4·4 | 3·0 | Active |
| 56 | m | 41 | n.a. | 1·5 | 1 | 1·5 | 70·9 | 4·5 | 3·3 | Active |
| 57 | f | 68 | 9 | 2 | 1 | 2 | 44·6 | 6·4 | 2·0 | Active |
| 58 | m | 64 | 11 | 4 | 4 | 4 | 10·0 | 0·8 | 0·5 | Remission |
| 59 | f | 46 | 12 | 2·5 | 3 | 3 | 10·5 | 1·9 | 1·8 | Remission |
| 60 | m | 49 | 6 | 2 | 2·5 | 2·5 | 15·0 | 1·7 | 1·3 | Remission |
| 61 | f | 49 | 14 | 3 | 3 | 3 | 3·2 | 2·6 | 1·1 | Remission |
| 62 | f | 57 | 25 | 4 | 4 | 4 | 15·0 | 1·4 | 1·1 | Remission |
| 63 | m | 48 | 8 | 2 | 1·5 | 1·5 | 16·5 | 2·6 | 1·0 | Remission |
| 64 | f | 61 | 14 | 3 | 3 | 3 | 16·2 | 4·1 | 0·9 | Borderline |
| 65 | f | 40 | 4 | 1·5 | 1·5 | 1·5 | 3·7 | 1·3 | 0·6 | Remission |
| 66 | f | 46 | 12 | 2·5 | 3 | 3 | 6·7 | 1·7 | 1·0 | Remission |
| 67 | m | 22 | 2 | 1 | 1 | 1 | 4·3 | 1·3 | 1·4 | Remission |
| 68 | m | 50 | 17 | 2 | 2 | 1·5 | 4·5 | 1·3 | 1·8 | Remission |
| 69 | f | 35 | 10 | 1·5 | 2 | 2·5 | 1·0 | 0·4 | 1·2 | Remission |
| 70 | f | 60 | 20 | 1·5 | 1 | n.a. | 6·5 | 1·1 | 1·5 | Remission |
| 71 | f | 58 | 23 | 2 | 2 | 2 | 6·2 | 1·6 | 0·8 | Remission |
| 72 | f | 44 | 10 | 2 | 2 | 2 | 10·1 | 6·2 | 1·1 | Active |
| 73 | m | 22 | 3 | 1 | 1 | 1 | 4·0 | 0·8 | 0·6 | Remission |
| 74 | f | 57 | 16 | 3·5 | 3·5 | 3·5 | 4·7 | 2·6 | 1·3 | Remission |
| 75 | f | 61 | 0·3 | 1·5 | 1·5 | 1·5 | 21·6 | 17·9 | 2·9 | Active |
| 76 | f | 28 | 1·3 | 2 | 2 | 2 | 2·98 | 0·9 | 2·1 | Remission |
| 77 | m | 47 | 0·3 | 2·5 | n.a. | n.a. | 7·9 | 1·0 | 1·0 | Remission |
| 78 | f | 32 | n.a. | 1·5 | n.a. | n.a. | 3·7 | 1·4 | 1·0 | Remission |
| 79 | f | 41 | 0·7 | 1 | 1 | 1 | 8·7 | 2·4 | 1·7 | Remission |
| 80 | f | 59 | 3 | 2 | 2·5 | 2 | 8·6 | 1·9 | 2·4 | Remission |
| 81 | f | 29 | 3 | 1 | 1 | 1 | 5·9 | 2·3 | 1·7 | Remission |
| 82 | m | 27 | n.a. | 1 | 1 | 1·5 | 13·9 | 3·3 | 1·1 | Remission |
| 83 | m | n.a. | n.a. | 4·5 | 4·5 | 4·5 | 3·2 | 0·8 | 0·6 | Remission |
| 84 | m | 35 | n.a. | 2·5 | 4 | n.a. | 4·4 | 1·1 | 2·1 | Remission |
| 85 | f | 38 | 3 | 2 | 2 | 2 | 14·7 | 4·9 | 2·1 | Active |
*Active disease versus remission was determined using the cut‐off values of the following indices: total FLC level = 17, κ M/D ratio = 4·0 and λ M/D ratio = 2·4. Active disease is supported when at least one of these 3 indices' values is above the cut‐off value. M = monomer; D = dimer; SP‐MS = secondary progressive MS; EDSS = expanded disability status scale; n.a. = data not available; m = male; f = female.
The control group included 28 healthy individuals (17 females). The age of control group individuals ranged from 20 to 60 years (mean age 38·7 ± 11·2) and was similar to that of MS patients. Healthy individuals showed no evidence for neurological, inflammatory or any other chronic disease; monoclonal gammopathy was also excluded. Saliva samples were collected from 28 healthy individuals; serum samples were collected from four healthy individuals.
Sample preparation
Saliva samples (approximately 0·5 ml) were collected during the morning hours (from 8 to 12 a.m.) by spitting, and stored in 1·5 ml Eppendorf tubes at −30°C until use. Before testing, the saliva samples were centrifuged at 16 000 g for 20 min in the Eppendorf centrifuge (Eppendorf 5415C; Marshal Scientific, Hampton, NH, USA). The supernatant aliquots were dried in a SpeedVac apparatus (SpeedVac Sc100; Savant, Farmingdale, NY, USA) and redissolved in electrophoresis sample buffer containing 0·2 mol/l sucrose, 6% sodium dodecyl sulphate (SDS), 125 mmol/l hydroxylmethylaminomethane (Tris), 4 mmol/l Na2 ethylenediamine tetraacetic disodium salt (EDTA). Equal volumes of the obtained samples were applied for their analysis by Western blot technique.
Western blotting
Western blotting was performed as described 3, 7. Briefly, samples were run by SDS‐electrophoresis, using high‐resolution 10–20% Nu‐Sep Tris‐Tricine gels (Gradipore, Frenchs Forest, New South Wales, Australia) under non‐reducing conditions. The electrophoretically separated proteins were blotted onto nitrocellulose membrane (Schleicher and Schuell, Dassel, Germany) using a Gradipore LongLife transfer buffer. FLC bands were immunodetected with rabbit antibodies to human Ig κ and λ light chains (Dako, Carpinteria, CA, USA). Immunodetection procedure was performed using the Bench Pro 4100 Card Processing station (Invitrogen Corp., Carlsbad, CA, USA), allowing automation of manual processing steps. Proteins were visualized with Super‐Signal West Pico Chemiluminescent Substrate (Pierce, Rockford, IL, USA) on X‐ray films (Fujifilm Corp., Tokyo, Japan).
Quantification of intensity of immunoreactive bands
Software 3, 7 was used to quantify the intensity of FLC immunoreactive bands (FLC monomers and covalently bound dimers). The obtained immunoreactivity values in the tested samples were normalized against that of a control sample, by dividing the intensity (I) of the tested samples by that of the control: Itested sample/Icontrol sample. The control sample comprised a mixture of saliva samples obtained from 23 of 28 healthy individuals participating in this study. A control sample was included in each electrophoretic run alongside the tested samples.
Total protein determination in saliva samples
Saliva supernatant aliquots were diluted (1 : 10) and analysed using the pyrogallol red–molybdate complex method 20 on Beckman Coulter analyser (Beckman Coulter, Inc., Brea, CA, USA).
Statistical analysis
Confidence intervals were calculated for the indices accounting for total FLC level, κ monomer/dimer (M/D) ratio and λ M/D ratio in active MS patients, MS patients in remission and healthy subject groups. The non‐parametric Mann–Whitney U‐test was also applied to compare the FLC indices in these three groups.
Results
Optimization of experimental conditions using normal saliva
Saliva samples of three healthy individuals were centrifuged, and the obtained supernatants were collected for further examination. To determine optimal sample dilution, saliva supernatant aliquots ranging between 0·9 and 30 μl were dried and then redissolved in equal amounts of electrophoresis sample buffer. Equal amounts of the prepared samples were subjected to Western blot analysis. Figure 1a demonstrates the relationship between chemiluminescence signal and saliva content in a sample obtained from one of the healthy individuals. Similar results were obtained by analysing saliva samples of two other healthy individuals. Figure 1b demonstrates the linearity range suitable for the FLC monomer and dimer measurements. We observed linearity of the chemiluminescence signal in a range from 0·015 to 0·06 that corresponded to the range of 5–15 μl of saliva supernatant used for sample preparation.
Figure 1.
Relationship between the chemiluminescence signal of free light chain (FLC) monomer (25 kDa) and dimer (50 kDa) bands and the amount (μl) of healthy saliva supernatant used for sample preparation. Axis x = saliva supernatant aliquots (μl) used for the preparation of samples; axis y = chemiluminescence signal. (a) Saliva supernatant aliquots of 0·93, 1·87, 3·75, 7·5, 15·0 and 30·0 μl were dried and redissolved in 45 μl electrophoresis sample buffer. Equal amounts of the prepared samples were applied for Western blot analysis (10 μl/well). The displayed electrophoregram demonstrates the increasing intensity of FLC bands in these samples (tracks 1–6, respectively). (b) Linear regression lines and coefficient of determination (R 2) values are displayed.
FLC analysis in healthy individuals
Figure 2 demonstrates Western blot analysis of the monomeric (25 kDa) and dimeric (50 kDa) κ and λ FLC in saliva of eight healthy subjects. Although the FLC monomer and dimer levels varied among different individuals, most of them (22 of 28) showed predominance of dimeric compared to monomeric FLC (κ and λ). In fact, the FLC monomers dominated in only six of the 28 healthy saliva samples.
Figure 2.
Western blot analysis of free light chains (FLC) in saliva samples from eight healthy subjects (tracks 2–9). Track 1 = control saliva sample. Predominance of FLC dimer (50 kDa) versus monomer (25 kDa) levels is observed.
FLC monomer–dimer patterns in saliva and serum samples of four healthy individuals were examined (Fig. 3). FLC patterns in serum differed considerably from those in saliva by showing either moderate or mild predominance of monomeric versus dimeric FLC. The observed patterns of FLC monomers and dimers in serum were in concordance with those obtained in our previous studies, where serum FLC patterns were examined in a larger number of healthy individuals (n = 30) 21, 22.
Figure 3.
Comparative Western blot analysis of free light chain (FLC) monomer–dimer patterns in saliva and serum samples of four healthy individuals (tracks 1–4). Saliva and serum samples, obtained from the same healthy individuals, were compared. Note the striking differences between saliva and serum samples with respect to the ratio of monomeric (25 kDa) to dimeric (50 kDa) FLC.
FLC indices for the total FLC level and for the M/D ratios of κ and λ FLC were estimated in saliva samples of 28 healthy individuals. The total FLC level indices varied from 1·3 to 14·1, the κ M/D ratio indices varied from 0·3 to 2·3 and λ M/D ratio indices varied from 0·2 to 2·2 (Table 5).
Table 5.
Saliva free light chain (FLC) monomer–dimer (M‐D) indices of healthy subjects (n = 28) *
| FLC M‐D indices | Range | Mean ± s.d. |
|---|---|---|
| κ M/D ratio | 0·3–2·9 | 1·4 ± 0·8 |
| λ M/D ratio | 0·2–2·2 | 0·8 ± 0·5 |
| Total FLC: λ (M+D) + κ (M+D) | 1·3–14·1 | 5·2 ± 3·9 |
To check whether food intake may affect the saliva FLC patterns of the same individual, we analysed saliva samples from five healthy subjects collected after overnight fasting and 1 h after the morning meal. Despite slight changes in FLC patterns before and after the meal (Fig. 4), the obtained values were within the normal ranges, as shown in Table 5.
Figure 4.
Western blot analysis of monomeric (25 kDa) and dimeric (50 kDa) free light chains (FLC) in saliva samples of four healthy individuals (HI1–HI4) before (tracks 1, 3, 5, 7) and after (tracks 2, 4, 6, 8) meal.
FLC patterns in MS
FLC monomer–dimer patterns in the saliva of active MS patients (n = 27, including 15 RR‐MS with relapse and 12 SP‐MS patients), MS patients in remission (n = 58) and healthy individuals (n = 28) were examined and compared. A large proportion of patients in remission (47 of 58) demonstrated FLC patterns comparable to those observed in healthy individuals (Fig. 5, tracks 2‐–5). In contrast to MS patients in remission, most active MS patients (24 of 27) tended to show either (a) an increase in the level of FLC monomers (κ or λ, or both) and/or (b) an increase in the total level of FLC (κ monomers plus dimers and λ monomers plus dimers). Figure 6 demonstrates FLC analysis in saliva of five active MS patients versus healthy individuals and versus patients in remission.
Figure 5.
Western blot analysis of free light chain (FLC) monomer–dimer patterns in saliva of five multiple sclerosis (MS) patients in remission (tracks 2–6). Track 1 = control saliva sample. Tracks 2–5 display normal monomer–dimer patterns of FLCs, in which dimers predominate. Track 6 shows the abnormal monomer–dimer pattern of κ‐FLC, evidenced by monomer preponderance.
Figure 6.
Western blot analysis of free light chain (FLC) monomers (25 kDa) and dimers (50 kDa) in saliva of five active multiple sclerosis (MS) patients (a2, b2, c2, d2, e2) versus healthy individuals (a1, b1) and versus patients in remission (c1, d2, e1). Active MS patients demonstrate abnormally increased intensity of the immunoreactive FLC bands and increased proportion of FLC monomers compared to healthy individuals and MS patients in remission.
As, in many cases, the FLC bands in saliva of active MS patients showed high chemiluminescence signals outside the linearity range, the samples were diluted further to allow valid quantification. An illustration of this approach is presented in Fig 7, which shows optimal sample dilutions suiting the linearity range.
Figure 7.
Relationship between the chemiluminescence signal of free light chain (FLC) monomer (25 kDa) and dimer (50 kDa) bands (axis y) and amount of saliva in sample (axis x) of patient with active multiple sclerosis (MS). Saliva supernatant aliquots of 0·156, 0·132, 0·625, 1·25, 2·5 and 5·0 μl were dried and redissolved in 45 μl electrophoresis sample buffer. Equal amounts of the prepared samples were applied for Western blot analysis (10 μl/well). The displayed electrophoregram demonstrates the increasing intensity of FLC bands of these samples (tracks 1–6, respectively). In this case, 1·25 μl of supernatant (used for sample preparation) was optimal for the intensity measurements within the linearity range.
The calculated FLC indices accounting for total FLC level and M/D ratios in saliva of active MS patients and patients in remission are presented in Tables 2, 3, 4.
FLC indices distinguishing a healthy state from MS
Mean values of saliva FLC indices of healthy individuals (Table 5) plus 3 standard deviations (s.d.) were computed and used as the cut‐off values which distinguish the healthy state from MS. The following cut‐off values of FLC indices were obtained: total FLC level index = 17, κ M/D ratio index = 4·0 and λ M/D ratio index = 2·4. All saliva samples of healthy individuals were below these cut‐off values (Fig. 8). In contrast, in 23 of 27 patients with active MS at least one of these three FLC indices was above the cut‐off value. Only three active MS patients (patients 16, 25 and 27) had normal FLC index values (below the cut‐off values), and FLC indices in one patient (patient 5) were borderline (Table 3 and Fig. 8). Of note, for the appropriate interpretation of the data obtained, both κ M/D and λ M/D indices should be considered. For example, patient 22 showed abnormalities in the κ M/D ratio index while the λ M/D ratio index was normal, whereas for patients 8 and 26 determination of λ M/D ratio indices was necessary to detect abnormalities in the FLC patterns (Fig. 8).
Figure 8.
Free light chain (FLC) analysis in saliva: (a) κ monomer/dimer (M/D) ratio index versus total FLC level index; (b) λ M/D ratio index versus total FLC level index. Saliva samples from healthy subjects (green), multiple sclerosis (MS) patients in relapse (red) and secondary progressive MS (SP‐MS) patients (black) are displayed. The cut‐off values are indicated by arrows on the x‐ and y‐axes. The numbered samples indicate cases, in which only one of the FLC indices, either κ M/D or λ M/D index, was abnormal.
Finally, the calculated saliva FLC indices of MS patients in remission (n = 58) were normal (below cut‐off values) in 43 patients, abnormally high in 11 patients and borderline in four patients (Table 4, Fig. 9).
Figure 9.
Free light chain (FLC) analysis in saliva: (a) κ monomer/dimer (M/D) ratio index versus total FLC level index; (b) λ M/D ratio index versus total FLC level index. Saliva samples from multiple sclerosis (MS) patients in remission (yellow), MS patients in relapse (red) and secondary progressive MS (SP‐MS) patients (black) are displayed. The cut‐off values are indicated by arrows on the x‐ and y‐axes.
FLC index values are statistically different in active MS, MS in remission and healthy state groups
The 95% confidence intervals of the FLC indices of patients with active MS, MS in remission and healthy subject groups were distinct (non‐overlapping) in all three groups with respect to total FLC level and λ M/D ratio indices. There was, however, some overlap between MS patients in remission and those with active disease regarding the κ M/D ratio (Table 6). Nevertheless, the Mann–Whitney U‐test showed that patients with active MS, MS patients in remission, as well as healthy subjects, differed significantly in all three parameters (total FLC level index, κ M/D ratio index and λ M/D ratio index) (Table 7). Finally, our technique differentiates between healthy state and active MS with a specificity of 100% and sensitivity of 88·5%, and between remission and active MS with a specificity 80% and a sensitivity of 89% (borderline values were excluded from the calculations).
Table 6.
Mean and the 95% confidence intervals of free light chain (FLC) indices in multiple sclerosis (MS) patients and healthy subjects
| Total FLC index | κ M/D ratio index | λ M/D ratio index | ||||
|---|---|---|---|---|---|---|
| Mean | Confidence interval | Mean | Confidence interval | Mean | Confidence interval | |
| Healthy individuals | 5·2 | 3·7–6·6 | 1·4 | 1·1–1·7 | 0·8 | 0·6–1·0 |
| MS patients in remission | 11·3 | 8·2–14·5 | 2·8 | 2·0–3·5 | 1·4 | 1·1–1·7 |
| Active MS patients | 31·6 | 17·4–45·8 | 6·2 | 3·2–9·1 | 8·6 | 3·8–13·4 |
M/D = monomer/dimer.
Table 7.
Comparison of free light chain (FLC) indices between multiple sclerosis (MS) patients and healthy subjects using the Mann–Whitney U‐test
| Total FLC index | κ M/D ratio index | λ M/D ratio index | |
|---|---|---|---|
| MS patients in remission versus healthy individuals | P = 0·0008 | P = 0·007 | P = 0·003 |
| MS patients in remission versus active MS patients | P < 0·0001 | P = 0·0002 | P < 0·0001 |
| Active MS patients versus healthy individuals | P < 0·0001 | P = 0·0002 | P < 0·0001 |
Discussion
In this study we have developed a new procedure based on a semiquantitative Western blot analysis to study FLC in saliva. The saliva samples of healthy subjects and those of MS patients were tested, and indices accounting for FLC monomer/dimer ratio and total FLC levels were calculated. The developed technique made it possible to discriminate between healthy state and active MS (specificity 100%, sensitivity 88·5%) and between active MS and remission (sensitivity 89%, specificity 80%). The developed procedure is non‐invasive, requires no expensive equipment and might be applied in clinical laboratories, as a new ancillary tool, to characterize the status of MS patients and their response to treatment.
To standardize the conditions of our technique, the saliva samples were collected and handled as equally as possible for all patients and control subjects. All saliva samples were centrifuged at the same speed to remove the cells and debris from the supernatant. As the salivary Ig levels may be affected by the rate of secretion and relation to meals 23, we checked the influence of fasting status on the FLC patterns. Although some minor changes in the FLC levels were observed, they were insignificant and did not affect the diagnostic capability of our technique.
Another important standardization issue was related to total FLC level index calculation. In our study all measurements were carried out using equal volumes of tested samples, i.e. without adjustment to total protein concentration. The rationale for such an approach was based on studies which showed that relating the immunoglobulin level to total protein might be misleading due to large individual variations of secretory response of salivary proteins 23, 24, 25. In addition, we have measured total protein concentration in 55 saliva samples (of 19 active MS patients, 22 patients in remission and 14 healthy individuals) by using the pyrogallol red–molybdate complex method 20 (data not shown). We found no significant correlation between total protein levels and total FLC level indices (two‐tailed P = 0·112; R 2 = 0·047 by Pearson's correlation test). Taken together, these findings refuted the possibility that increased values of total FLC index might be due to total protein increase.
Comparison of the FLC patterns of normal saliva and normal serum revealed striking differences, showing a predominance of dimers in the saliva samples and monomers in the sera. We hypothesize that this difference stems from the structural differences of the Ig contained in the saliva and in the serum. While the saliva Ig are made mainly of the locally synthesized IgA1 and IgA2 isotypes, the serum Ig are mainly IgG. Further, in contrast to other Ig classes, the IgA2 heavy chains are bound to dimeric, not monomeric, light chains 26, therefore production of a larger amount of dimeric light chains may be needed for the biosynthesis of intact IgA2 molecules. Thus, most probably, the differences in FLC monomer–dimer patterns observed in normal saliva versus serum are related to a higher proportion of the IgA2 synthesizing B cells in oral mucosa which produce larger amounts of dimeric light chains, compared to the peripheral B cells synthesizing other Ig molecules.
Our study found that most patients with active MS undergo changes in their saliva FLC patterns, which show a higher proportion of monomeric FLC, as follows from their κ M/D or λ M/D indices (Tables 2 and 3). The reason for this change still needs to be determined, but it could be speculated that it is due to a local switch in the biosynthesis from IgA2 to IgA1 or IgG, leading to increased production of monomeric FLC. Alternatively, it may be due to penetration of peripheral B lymphocytes to oral mucosa and production of larger amounts of monomeric FLC.
The non‐overlapping confidence intervals of saliva FLC indices (Table 6) suggest that FLC patterns of MS patients in remission place them between normal individuals and MS patients with active disease. This finding lends further support to the view that even during clinical remission there is a subtle immunological activity, which may need special attention and perhaps certain measures for suppression. Fifteen of 58 patients in remission continued to have higher than normal FLC indices, thus suggesting ongoing immunological activity despite perceived clinical remission. It is already known that MS flare may be preceded by T cell and B cell activity months before the clinical flare 27, 28. Whether our finding has treatment implications still needs to be investigated.
It is also worth mentioning that four of 27 active MS patients had FLC indices lower than the cut‐off values, and thereby failed to fulfil the saliva FLC criteria of active disease. Except for patient 5, who had borderline FLC indices, the other three patients were of the SP‐MS type and had high EDSS scores ranging from 4·5 to 7·5 (Table 3). As the immunopathological mechanisms underlying the SP and RR‐MS forms are different 29, 30, analysis of a larger number of MS cases (RR‐MS with relapse and SP‐MS) is needed to compare the diagnostic efficacies of our methodological approach in these two distinct subsets of active MS.
In conclusion, we have developed a new technique to analyse monomeric and dimeric FLCs in the saliva of MS patients. Application of this technique revealed deviation from the normal FLC monomer–dimer profile in the saliva of patients with active MS, while these pathological changes were absent in a large number of MS patients in remission. We think that our technique offers an easy, simple and non‐invasive way to follow the disease course. Further studies, including longitudinal monitoring of the saliva FLC profiles in the individual MS patients, might be helpful in therapeutic decisions and in generation of new insights on the immunopathology of MS.
Disclosure
The authors have no financial and commercial conflicts of interest to disclose.
Acknowledgements
B. K. and S. G. performed the experiments. E. G.‐C., A. M. and E. K. collected the samples and provided clinical information. B. K., S. G. and E. G.‐C. designed the study. B. K., S. G., E. G.‐C., I. B.‐Z., A. L. and S. F. participated in discussions and interpretation of the obtained results. B. K., A. L. and S. F. wrote the manuscript. We would like to thank Dr R. Kaplan for performing the statistical analyses and for his invaluable advice and helpful discussion in the preparation of this manuscript and Dr R. Yeskaraev for performing total protein analysis. The study was supported by the National Multiple Sclerosis Society, grant no. PP‐1512–07139.
[Correction added on 26 January 2018, after first online publication: Addition for the author E. Ganelin‐Cohen affiliation has been reflected in this version.]
References
- 1. Fischer C, Arneth B, Kehler L, Lotz J, Lackner KJ. Kappa free light chains in cerebrospinal fluid as markers of intrathecal immunoglobulin synthesis. Clin Chem 2004; 50:1809–13. [DOI] [PubMed] [Google Scholar]
- 2. Presslauer S, Milosavljevic D, Brucke T, Bayer P, Hubl W. Elevated levels of kappa free light chains in CSF to support the diagnosis of multiple sclerosis. J Neurol 2008; 255:1508–14. [DOI] [PubMed] [Google Scholar]
- 3. Kaplan B, Aizenbud BM, Golderman S, Yaskariev R, Sela BA. Free light chain monomers in the diagnosis of multiple sclerosis. J Neuroimmunol 2010; 229:263–73. [DOI] [PubMed] [Google Scholar]
- 4. Sela B, Kaplan B, Aizenbud B. A new laboratory method for the diagnosis of multiple sclerosis. J Isr Neurol Assoc 2010; 3:28–30. [Google Scholar]
- 5. Kaplan B, Livneh A, Sela BA. Immunoglobulin free light chain dimers in human diseases. Sci World J 2011; 11:726–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kaplan B, Golderman S, Sela B. Immunoglobulin free light chains in the diagnosis of multiple sclerosis In: Scholz E, Muller C, eds. Multiple sclerosis: causes, diagnosis and management. New York: Nova Science Publishers, 2012:131–42. [Google Scholar]
- 7. Kaplan B, Golderman S, Yahalom G et al Free light chain monomer‐dimer patterns in the diagnosis of multiple sclerosis. J Immunol Methods 2013; 390:74–80. [DOI] [PubMed] [Google Scholar]
- 8. Desplat‐Jégo S, Feuillet L, Pelletier J, Bernard D, Chérif AA, Boucraut J. Quantification of immunoglobulin free light chains in cerebrospinal fluid by nephelometry. J Clin Immunol 2005; 25:338–45. [DOI] [PubMed] [Google Scholar]
- 9. Presslauer S, Milosavljevic D, Huebl W et al Validation of kappa free light chains as a diagnostic biomarker in multiple sclerosis and clinically isolated syndrome: a multicenter study. Mult Scler 2016; 22:502–10. [DOI] [PubMed] [Google Scholar]
- 10. Passerini G, Dalla Costa G, Sangalli F et al Free light chains and intrathecal B cells activity in multiple sclerosis: a prospective study and meta‐analysis. Mult Scler Int 2016; 2016:1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Thio M, Blokhuis BR, Nijkamp FP, Redegeld FA. Free immunoglobulin light chains: a novel target in the therapy of inflammatory diseases. Trends Pharmacol Sci 2008; 29:170–4. [DOI] [PubMed] [Google Scholar]
- 12. Franciotta D, Salvetti M, Lolli F, Serafini B, Aloisi F. B cells and multiple sclerosis. Lancet Neurol 2008; 7:766–7. [DOI] [PubMed] [Google Scholar]
- 13. Coyle PK, Bulbank M. Immune‐reactive cells in multiple sclerosis mucosal secretions. Neurology 1989; 39:378–80. [DOI] [PubMed] [Google Scholar]
- 14. Coyle PK. Molecular analysis of IgA in multiple sclerosis. J Neuroimmunol 1989; 22:83–92. [DOI] [PubMed] [Google Scholar]
- 15. Nouri M, Bredberg A, Weström B, Lavasani S, Lees JR. Intestinal barrier dysfunction develops at the onset of experimental autoimmune encephalomyelitis, and can be induced by adoptive transfer of auto‐reactive T cells. PLOS ONE 2014; 9:e106335. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Calais G, Forzy G, Crinquette C et al Tear analysis in clinically isolated syndrome as new multiple sclerosis criterion. Mult Scler 2010; 16:87–92. [DOI] [PubMed] [Google Scholar]
- 17. Coyle PK, Sibony PA, Johnson C. Increased monomeric immunoglobulin A levels in tears from multiple sclerosis patients. Ann Neurol 1987; 21:211–4. [DOI] [PubMed] [Google Scholar]
- 18. Devos D, Forzy G, De Seze J et al Silver stained isoelectrophoresis of tears and cerebrospinal fluid in multiple sclerosis. J Neurol 2001; 248:672–5. [DOI] [PubMed] [Google Scholar]
- 19. Polman CH, Reingold SC, Banwell B et al Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011; 69:292–302. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Fujita Y, Mori I, Kitano S. Color reaction between pyrogallol red–molybdenum (VI) complex and protein. Bunseki Kagaku 1983; 32:E379–86. [Google Scholar]
- 21. Kaplan B, Golderman S, Aizenbud B et al Immunoglobulin‐free light chain monomer‐dimer patterns help to distinguish malignant from premalignant monoclonal gammopathies: a pilot study. Am J Hematol 2014; 89:882–8. [DOI] [PubMed] [Google Scholar]
- 22. Kaplan B, Ramirez‐Alvarado M, Sikking L et al Free light chains in plasma of patients with light chain amyloidosis and non‐amyloid light chain deposition disease. High proportion and heterogeneity of disulfide‐linked monoclonal free light chains as pathogenic features of amyloid disease. Br J Haematol 2009; 144:705–15. [DOI] [PubMed] [Google Scholar]
- 23. Brandtzaeg P. Secretory immunity with special reference to oral cavity. J Oral Microbiol 2013; 5:2041–65. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Brandtzaeg P, Berdal P. J chain in malignant human IgG immunocytes. Scand J Immunol 1975; 4:403–7. [DOI] [PubMed] [Google Scholar]
- 25. Rudney JD, Smith QT. Relationships between levels of lysozyme, lactoferrin, salivary peroxidase, and secretory immunoglobulin A in stimulated parotid saliva. Infect Immun 1985; 49:469–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Grey HM, Abel CA, Yount WJ, Kunkel HG. A subclass of human γA globulins (γA2) which lacks the disulfide bonds lining heavy and light chains. J Exp Med 1968; 128:1223–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Mills EA, Mirza A, Mao‐Draayer Y. Emerging approaches for validating and managing multiple sclerosis relapse. Front Neurol 2017; 8:116. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Hohmann C, Milles B, Schinke M et al Categorization of multiple sclerosis relapse subtypes by B cell profiling in the blood. Acta Neuropathol Commun 2014; 2:138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Christensen JR, Börnsen L, Ratzer R et al Systemic inflammation in progressive multiple sclerosis involves follicular T‐helper, Th17‐ and activated B‐cells and correlates with progression. PLOS ONE 2013; 8:e57820. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Blauth K, Owens GP, Bennett JL. The ins and outs of B cells in multiple sclerosis. Front Immunol 2015; 6:565. [DOI] [PMC free article] [PubMed] [Google Scholar]