This manuscript describes the use of nasal and bronchial absorption techniques, specifically using synthetic absorptive matrices (SAM) to sample the mucosal lining fluid (MLF) of the upper and lower airway. These methods provide better standardization and tolerability than existing respiratory sampling techniques.
The methods of nasal absorption (NA) and bronchial absorption (BA) use synthetic absorptive matrices (SAM) to absorb the mucosal lining fluid (MLF) of the human respiratory tract. NA is a non-invasive technique which absorbs fluid from the inferior turbinate, and causes minimal discomfort. NA has yielded reproducible results with the ability to frequently repeat sampling of the upper airway.
By comparison, alternative methods of sampling the respiratory mucosa, such as nasopharyngeal aspiration (NPA) and conventional swabbing, are more invasive and may result in greater data variability. Other methods have limitations, for instance, biopsies and bronchial procedures are invasive, sputum contains many dead and dying cells and requires liquefaction, exhaled breath condensate (EBC) contains water and saliva, and lavage samples are dilute and variable.
BA can be performed through the working channel of a bronchoscope in clinic. Sampling is well tolerated and can be conducted at multiple sites in the airway. BA results in MLF samples being less dilute than bronchoalveolar lavage (BAL) samples.
This article demonstrates the techniques of NA and BA, as well as the laboratory processing of the resulting samples, which can be tailored to the desired downstream biomarker being measured. These absorption techniques are useful alternatives to the conventional sampling techniques used in clinical respiratory research.
Most respiratory diseases cause an inflammatory response, and there is an urgent need for sampling from the respiratory mucosal surface in allergic rhinitis, viral and fungal infections, tuberculosis, asthma, chronic obstructive pulmonary disease, pulmonary fibrosis and lung cancer 1. Nasopharyngeal aspirate (NPA), nasal lavage, and bronchoalveolar lavage (BAL) are common techniques for sampling the upper and lower airway. However, these techniques present considerable problems including poor tolerability, dilution of inflammatory mediators, and the inability to frequently repeat sampling2. One alternative to NPA sampling is the use of swabs, either nylon flocked, cotton, or rayon3,4, but these also have limitations, since they cause discomfort and damage to the nasal epithelium, and in some cases irreversible binding of inflammatory mediators5. These techniques cannot generally be repeated serially over an hour, and alternative techniques may be more effective for detection of low abundance cytokines and chemokines5,6. Additionally, the user variability associated with these techniques may generate inconsistencies in data, resulting in the requirement of larger patient cohorts.
Alternatively, absorption techniques using both natural and synthetic sponges have been used to collect MLF from mucosal surfaces. Ophthalmic sponges composed of natural cellulose (e.g., Weck-cel) have been used to sample saliva, cervical, and vaginal secretions7. In addition, synthetic sponges made from polyvinyl alcohol (PVA) and hydroxylated polyvinyl acetate (HPVA) have been utilised8. Seven different absorptive materials have been compared for sampling oral fluid prior to measuring antibodies9, while polyurethane mini-sponges have been used to collect human tears10.
Filter paper consisting of natural cellulose from the cotton plant has been widely used to absorb nasal secretions since the pioneering paper of Alam and colleagues in 199211,12,13,14,15,16,17. Filter paper discs have been produced from filter cards (e.g. Shandon), and have been utilized to measure histamines and cytokines after controlled nasal allergen challenges and with natural allergen exposure18,19,20,21. However, different batches of filter paper vary in their degree of protein binding and some fail to release cytokines. Methods using a synthetic absorptive matrix (SAM) have therefore been developed2,22,23. SAMs are now generally used to obtain nasal MLF by NA. These absorbent materials are comfortable to use and can obtain MLF even from inflamed noses at frequent intervals over extended periods of time.
Nasal absorption is a form of Precision Mucosal Sampling using a SAM for the sampling of MLF in the upper airway. NA devices are manufactured as CE-marked medical devices from medical grade materials using clean rooms and are free of dust and allergens. The NA sampler consists of a handle and SAM in a sterile cryotube. The SAM consists of polymers, typically fibers, but it is also available as foam, and these are selected to be soft and absorptive, with rapid wicking for sample collection. SAMs have minimal protein binding to allow the efficient elution of absorbed secretions. NA is a very gentle, non-invasive technique that can be performed on donors of all ages. In addition, serial sampling, even every few minutes, is possible. NA has been validated against existing upper airway sampling techniques5 and repetitive sampling has allowed generation of kinetic data following challenge of the airway with allergen23,24,25, bacterial endotoxin26 and viral-type TLR agonists (Jha, A. et al., manuscript in preparation). NA has also been used in infants to investigate the natural history of atopy27,28,29 and in viral bronchiolitis30.
Bronchoscopic microsampling (BMS) is a procedure for collection of MLF in the lower airway that has been developed by Olympus31,32,33. Unfortunately, this BMS system is only licensed in Japan. Olympus supply two BMS systems: one with a fibrous hydroxylated polyester (FHPE) probe34,35,36,37, and one with a cotton probe33,38,39,40,41,42,43. A major stumbling block has been that the BMS probe used in patients with asthma caused mucosal contact bleeding, with half of all samples contaminated with blood. The authors concluded that it was not feasible to sample MLF using this BMS system from peripheral airways in asthma patients43.
As an alternative, we have developed BA using a soft SAM that can be performed during bronchoscopic investigation of the lower airways, including following experimental infection of asthmatic subjects with rhinovirus6. The BA device consists of: an external hollow catheter, a hand-piece that on activation extrudes the SAM, and a central plastic guide wire that has the SAM on its end. As for NA, BA kits are manufactured from medical grade materials using clean rooms and are free of dust and allergens. Additionally, devices are CE marked and are provided gamma-irradiated. The SAM strip is soft, absorptive, and has rapid wicking for sample collection. It also has minimal protein binding to allow the efficient elution of absorbed secretions. The device can fit through the working channel of a bronchoscope and can be used to rapidly and accurately sample MLF at specific sites of interest within the airway. Unlike BAL or BMS, BA does not result in contact bleeding or additional patient discomfort post-procedure.
Careful consideration should be given to the processing of NA and BA samples. Samples can be directly frozen and processed in batches, or can be processed immediately. The type of processing can be tailored toward certain downstream applications, including immunoassays for cytokines, chemokines and immunoglobulins, or elutions of viral, bacterial, and host cell associated RNA. We present the clinical collection and laboratory processing techniques associated with NA and BA as a guide for clinical researchers.
The techniques used in the following protocol have been approved by the West London Research Ethics Committee (Reference number 15/lO/0444).
1. Nasal Absorption (NA)
2. Bronchial Absorption (BA)
3. Processing of NA and BA Samples
Note: There are numerous options for laboratory processing of samples resulting from NA and BA. These protocols seek to store samples for later use, and to elute the MLF sample from the SAM.
NA has been utilized in a number of studies to easily and non-invasively measure mucosal inflammation. Following administration of allergen to the nose, prostaglandin-D2 (PGD2) levels can be observed to rise and fall within minutes (Figure 1), in line with mast cell degranulation (Thwaites et al., manuscript in preparation). Additionally, mediators of type-II inflammation, such as IL-5 (Figure 2, reprinted with permission from 24), can be measured in the hours following nasal allergen challenge23,24. In experimental infection of allergic asthmatics and healthy controls, NA was used to measure a panel of mediators, including interferon-gamma (IFN-γ), over the course of 7 days (Figure 3, reprinted with permission from 6). Additionally, in natural respiratory syncytial virus (RSV) infection of infants, NA demonstrated RSV to be associated with elevated levels of inflammatory cytokines, such as IFN-γ, relative to non-RSV infants with bronchiolitis and healthy controls (Figure 4, reprinted with permission from 30). Interestingly, this discrimination between RSV and non-RSV bronchiolitis was not significant in time-matched NPA samples (Figure 4).
BA was also used during experimental infection of allergic asthmatics with rhinovirus. At day 4 of rhinovirus infection, levels of IFN-γ, CXCL11, IL-10, and IL-5 were elevated from baseline (Figure 5; A, B, C and D, respectively). Additionally, this technique demonstrated elevated IL-5 levels in the lower airway of allergic asthmatics during rhinovirus infection, relative to healthy controls (Figure 5D)(Figure 5 reprinted with permission from 6).
These representative results were generated from samples eluted using assay buffer containing 0.05% Tween-20 and 1% BSA (see Table of Materials).
Figure 1: Rapid generation and clearance of prostaglandin-D2 following nasal allergen challenge. Levels of prostaglandin-D2 (PGD2) measured from nasal absorption eluates in serial samples following nasal allergen challenge with Timothy grass pollen (n=5). Each line represents data from one individual. Please click here to view a larger version of this figure.
Figure 2: Kinetic measurement of IL-5 following nasal allergen challenge. Production of IL-5 in serial nasal absorption samples following nasal allergen challenge. Three repeat allergen challenge studies were conducted, with the participants (n=19) receiving placebo (blue), low dose (10 mg) oral prednisone (orange), or high dose (25 mg) oral prednisone (red) one hour prior to allergen administration. Lines denote geometric mean and error bars are 95% confidence intervals of all participants. (Figure reprinted with permission from Leaker et al., Mucosal Immunology, 2017). Please click here to view a larger version of this figure.
Figure 3: Induction of Interferon-γ during rhinovirus infection. During infection challenge of healthy adults (n= 11, blue) and allergic asthmatics (n= 28, red) with rhinovirus-16, nasal absorption sampling was used to measure levels of interferon-γ (IFN- γ). Data are represented as A) raw spaghetti plots of individuals and B) median levels with error bars denoting interquartile ranges. (Figure reprinted with permission from Hansel et al.6). Please click here to view a larger version of this figure.
Figure 4: Nasosorption discriminates elevated Interferon-γ associated with RSV infection of infants. Nasal absorption and nasopharyngeal aspiration (NPA) were used to measure interferon-γ levels in infants with bronchiolitis associated with respiratory syncytial virus (RSV) infection (red, n=12), a non-RSV respiratory pathogen (green, n=12), and healthy controls (blue, n=9). Data were analyzed using a Kruskall-Wallis test with Dunns correction for multiple comparisons (***p<0.001). Lines denotes median values and error bars are interquartile ranges. (Figure modified with permission from Thwaites et al.30). Please click here to view a larger version of this figure.
Figure 5: Inflammatory mediators in bronchial absorption samples during rhinovirus infection.
Following rhinovirus-16 infection of healthy controls (n=10, blue) and allergic asthmatic volunteers (n=23, red), bronchial absorption was used to measure levels of A) IFN-γ, B) CXCL11, C) IL-10, and D) IL-5 at baseline and on day 4 of infection. Data analyzed by Wilcoxon signed rank test (matched samples) and Mann-Whitney test (unmatched samples). Figure reprinted with permission from Hansel et al.6 Please click here to view a larger version of this figure.
Results from existing airway sampling techniques are regarded as highly variable; alternative sampling techniques are needed to standardize research within this field5. NA and BA permit sampling of MLF in a non-invasive manner, and have exciting potential to measure immune responses in healthy and diseased airways. These techniques offer numerous potential advantages over existing techniques, including greater tolerability, speed of sampling, the ability to frequently repeat sampling, lower inter-user variability, and decreased dilution of immune mediators5. The duration of absorption and the processing technique used should be optimized for each study and rigorously maintained between sampling events. Additionally, in the case of BA, the site of sampling within the airway should be carefully replicated between individuals.
NA, and particularly BA, are still relatively novel techniques for clinical research. However, the benefits of these techniques have resulted in their use in numerous studies, including careful validation against alternative techniques5. These devices are now available as CE-marked devices ready for widespread use in respiratory research. While NA and BA result in much smaller sample volumes than alternative sampling techniques, higher obtained concentrations can result in greater sensitivity for low abundance immune mediators.
Depending on desired downstream applications, NA and BA samples can be directly frozen for later processing, enhancing study feasibility in a clinical research environment. The protocol for sample handling can also be adapted to suit particular downstream applications. The suggested processing techniques can be used for collection of protein or lipid immune mediators or nucleic acids, but should be optimized for each study. In particular, MLF can be eluted with different buffers. Firstly, immunoassay buffer can be used to measure mucosal cytokines, chemokines, and antibodies6,45. Buffers with higher detergent levels can also be used to guarantee cell lysis occurs, allowing the inclusion of intra-cellular cytokines. Chaotropic RNA extraction buffers should be used for the determination of viral infection, viral load, host mRNA, and the microbiome. Alternatively, organic solvents can be used for lipidomics and mass spectrometry.
In conclusion, direct absorption of MLF from mucosal surfaces is an exciting technique with potential use in respiratory, gastro-intestinal, urogenital, and other mucosal diseases. However, these promising absorption techniques will require precise validation of sampling and processing technique for individual assays (biomarkers) in each disease setting. In addition, these novel precision mucosal sampling techniques will require validation against conventional samples, such as from blood, breath, and sputum. With these techniques, MLF can be used to measure microbes, cytokines, chemokines, prostanoids, and antibodies.
The authors have nothing to disclose.
Funding: This work was supported by funding from the Imperial National Institute for Health Research (NIHR) Biomedical Research Centre (BRC), the NIHR Health Protection Research Unit (HPRU) in Respiratory Infections at Imperial College London in partnership with Public Health England (PHE) and the NIHR Imperial Patient Safety Translational Research Centre. The views expressed are those of the authors and not necessarily those of the NHS, the NIHR, the Department of Health or Public Health England.
Nasosorption (adult, 7mm width) | Hunt Developments | NSFL-FXI-11 | Different sizes are available for different patient groups/ages. |
Bronchosorption | Hunt Developments | BSFL-FXI-11 | Minimum bronchoscope channel size 2mm; Max working length 815mm |
Corning Costar Spin-X centrifuge tube filters (without membrane) | Sigma Aldrich | CLS9301-1000EA | |
Corning Costar Spin-X centrifuge tube filters (0.22um membrane) | Sigma Aldrich | CLS8160-24EA | For sterilisation of samples with infection risk. |
Assay (elution) buffer | Millipore | AB-33k | Not listed on the Millipore website but available through enquiry or general lab supply companies, such as Cedarlane. Contains 0.05% Tween-20 and 1% BSA. |
NP-40 Cell lysis buffer | Life Technologies | FNN0021 | Add bovine serum albumin to 1% (w/v). Can recover higher absolute mediator levels. |
Buffer RLT (RNA extraction) | Qiagen | 79216 | Allows recovery of RNA from nasosorption and bronchosorption samples. |
Trifluoroacetic acid | Sigma Aldrich | 302031-100ML-M | For elution of samples to be used in HPLC applications |
2.0ml micro-centrifuge tubes | Costar | 3213 | 2ml tubes are required for the Spin-X tube filters, traditional 1.5ml tubes will not fit these. |