INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Monitoring or analysis of humoral immune components in conventional body fluids, such as blood and urine, involves common methods. However, measurement of these factors in nonconventional biological fluids, such as cervical secretions, vaginal fluid, and saliva, is complex. The difficulties lie not only in the analysis of these mucosal fluids but also in obtaining reproducible and unaltered samples. The accuracy and consistency of the sampling procedures can ultimately affect experimental outcomes and quantitation of the individual components. Yet the assessment of such fluids is important because it gives insight into the local immune response, physiological modifications induced by infection, and potential drug profiles at the site of action (11, 12).

There are a number of different methods available for the collection of genital tract secretions, such as cervicovaginal washes, aspiration, and Wick absorption. Each technique has proven its utility but also has a downside. The washes yield a significant amount of material but combine vaginal and cervical secretions. These two secretions have different roles in the protection of the genital tract, and combining them prevents studies of each secretion. Aspiration works well for collection of cervical secretions at midcycle in women who are ovulating but yields little volume at other times in the cycle or when women are using oral contraceptives. Finally, the Wicks collect cervical and vaginal secretions individually but absorb a very small amount of material. The collection volume limits the number of possible analyses that can be performed on each sample. To overcome some of these problems, to standardize the methods of collection across different types of mucosal secretions, and to simplify the collection process in order to incorporate collection of secretions into multicenter clinical trials, the Weck-Cel method using ophthalmic sponges was developed.

Ophthalmic sponges were used successfully for the collection of secretions from the oral and genital tract mucosae to measure antibody levels in response to vaccination (3, 9). The consistency of immunoglobulin (Ig) recovery from the sponges was demonstrated previously (3). This collection technique overcomes some of the limitations encountered when washes or aspiration is used to obtain secretions, and a single device can be used for the collection. The Weck-Cel sponges were designed for the collection of tears and easily absorb fluids without causing trauma to the cervix or local tissue. Also, the nonabrasive collection does not interfere with Pap smear results (6). This method allows for assessment of a dilution factor for each individual secretion collected, therefore reducing the variability induced by unknown dilution of the samples (7). The cellulose fibers in the sponges are highly absorbent and have a low binding affinity for protein. Finally, this technique is simple and the procedure can be completed within 2 min, allowing easy incorporation into clinical trials.

To expand our understanding of immunoregulation in the genital tract, a study of cytokine concentrations using the Weck-Cel sponge for collection of cervical secretions from women was undertaken. However, during those studies it was discovered that some cytokines, unlike Ig, bind to the sponges, thereby preventing the diffusion of individual cytokines out of the sponges during the extraction procedure. These studies were conducted to optimize the processing procedure to ensure consistent release of individual cytokines and Ig from the sponges. Using this technique, baseline concentrations of cytokines in cervical secretions of women using oral contraceptive pills were determined. The objective of these studies was to demonstrate the utility of this method for collection of genital tract secretions throughout the menstrual cycle and to establish consistency in the volume of material and the recovery of each cytokine obtained from the sponges.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Immunochemical reagents. Affinity-purified F(ab')₂ fragments of goat antibodies specific for human IgG and IgA were purchased from Jackson Immunoresearch (West Grove, Pa.). All secondary antibodies were biotinylated affinity-purified F(ab')₂ fragments of goat anti-human IgG or IgA purchased from BioSource International (Camarillo, Calif.). Recombinant cytokine enzyme-linked immunosorbent assay (ELISA) kits were obtained from BioSource International. The kits used were for the cytokines interleukin 1 (IL-1), IL-2, IL-4, IL-5, IL-8, gamma interferon (IFN-, and granulocyte-macrophage colony-stimulating factor (GM-CSF). High-sensitivity ELISA kits were used for detection of IL-6, IL-10, and IL-12.

Cytokine extraction method. Recombinant cytokines obtained from BioSource International as part of the standard curve were utilized for the extraction and recovery experiments. The Weck-Cel sponges were allowed to absorb a solution containing known concentrations of cytokine and Ig calibrator (Binding Site, Birmingham, United Kingdom). The following are the respective cytokine concentrations (in picograms per milliliter) absorbed onto the sponges: IL-1, 3,000.0; IL-2, 1,562.5; IL-4, 312.5; IL-5, 1,250.0; IL-6, 77.5; IL-8, 2,500; IL-10, 1,250.0; IL-12, 500.0; GM-CSF, 1,500.0; and IFN-, 250.0. Cytokine and serum standards were diluted in the standard diluent provided by BioSource International for their ELISA kits, and 1% fetal calf serum (FCS) was added to the final solution. The standard diluent and FCS served as negative controls and were tested for each cytokine. In order to optimize the recovery of cytokines or Ig (see Results), the sponges underwent various treatment. Percent recovery was based upon a comparison of the absolute amount of cytokine in the extracted material compared to a stock solution that was not subjected to the extraction procedure.

Total Ig and cytokine concentration analysis. Ig concentrations were determined by quantitative ELISA as previously described (8). Microtiter plates were coated overnight at 4°C with affinity-purified antibodies specific for the appropriate isotype diluted in PBS. Total Ig assays were standardized using an Ig calibrator (Binding Site) of known IgA and IgG concentrations. The extracted samples and the calibrator were diluted in PBS containing 1% FCS (PBS-FCS) and added to the plates in duplicate. After an overnight incubation at 4°C the plates were washed with PBS with 0.05% Tween 20, isotype-specific biotinylated antibodies were added, and the mixture was incubated for 2 h at room temperature. The plates were then washed with PBS-0.05% Tween 20 and incubated for 30 min with horseradish peroxidase-conjugated avidin (Sigma, St. Louis, Mo.), diluted to 0.5 µg/ml, and 0.87% saline containing 0.05% Tween 20. The plates were washed and exposed to a substrate consisting of 2,2'-azino-bis(3-ethylbenzthiazoline-6-sulfonic acid) (Sigma) at a concentration of 0.25% (wt/vol) in 0.1 M citrate-phosphate buffer (pH 4.2) containing 0.0075% H₂O₂. Substrate turnover was monitored spectrophotometrically at 415 nm with an automated reader (BioTek, Winooski, Vt.). A standard curve was produced with Delta Soft ELISA analysis software.

Collection of biological specimens. Seven healthy women (age range, 19 to 35 years) who had taken OrthoCept oral contraceptive pills for a minimum of 2 months were enrolled in this study, approved by the Institutional Review Board at Magee-Womens Hospital. All women had normal Pap smears and no evidence of infection during a pelvic exam. Saliva, cervical secretions, and vaginal secretions were collected with Weck-Cel sponges. Cervical secretions were collected first after exposure of the cervical os with the speculum. The secretions were collected by placing the ophthalmic sponge directly into the cervical os and allowing it to absorb secretions for approximately 1 min. Vaginal secretions were collected by placing the ophthalmic sponge against the vaginal wall, and in a similar fashion, saliva was collected by placing the ophthalmic sponge over the parotid duct and allowing the sponge to absorb saliva.

Clinical specimen analysis. To determine the volume of secretions absorbed into the sponges, the final optimized method developed through these studies was used. Each individual sponge was weighed. The sponges were then equilibrated in 300 µl of extraction buffer for 30 min at 4°C. The diluted secretions were separated from the sponge using a Spin-x centrifuge filter unit (Costar) spun at 16,000 × g for 15 min at 4°C. Another 300 µl of buffer was added to each spear, and centrifugation at 16,000 × g for 15 min was performed immediately. In calculating the final concentration of immune components measured in the secretions, a dilution factor was determined based on the following formula: dilution factor = [(x0.06 g) + 0.6 g of buffer]/(x0.06 g), where x equals the weight of the sponge after collection and 0.06 g is the weight of the dry spear. The density of the buffer is 1.005 g/ml. Therefore, by conversion, 0.6 ml of buffer is 0.603 g. Given that x is the weight (in grams) of the sponge after collection, the dilution factor can be determined as a dimensionless number. (Note that the weight of the dry sponge is dependent on the lot of the sponges; therefore, the dry weight of the sponge was determined by measuring 10 sponges in each lot. This weight will vary ±3%.) This dilution factor was used to calculate the final concentration of cytokines and Ig in the secretions. Extracted samples were tested for blood content by using Hemastix strips (VWR Scientific Products, Bridgeport, N.J.). Any samples with blood contamination were discarded and not analyzed.

Statistical analysis. Extraction results were analyzed with Statview software. Differences between concentration measurements observed at days 9 and 20 of the menstrual cycle were determined by nonparametric analysis using the Mann-Whitney test. Significance was set at a P value of 0.05.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

Optimization of the extraction process. The extraction buffer used to separate the protein components of the secretions from the surface and internal matrix of the sponge consisted of PBS, 100 µg of the protease inhibitor aprotinin per ml, 0.1% sodium azide, and 0.25 M sodium chloride. During development of the extraction process, various concentrations of sodium chloride were tested for effects on extraction of Ig. However, increasing the concentration of NaCl beyond 0.25 M did not enhance recovery of cytokines from sponges. Another component of the buffer in the original protocol was FCS, which was initially added to the extraction buffer to aid in release of protein. However, during the analysis of clinical specimens, the FCS interfered with the measurement of blood contamination in cervical secretion specimens. To determine if FCS was necessary for release of protein, a series of triplicate sponges were loaded with IgA and extracted; 3.6 ± 0.2 and 3.5 ± 0.2 µg/ml (mean ± standard deviation) were obtained from the samples with and without FCS in the extraction buffer, respectively. No significant difference between the two buffers was found. In the final protocol FCS was added to the samples after measurement of blood contamination.

Cytokines in cervical secretions. To establish normal concentrations of cytokines in cervical secretions and demonstrate the usefulness of the described collection method, clinical samples from seven women were analyzed to determine the concentrations of IL-5, IL-6, IL-10, IL-8, IL-12, and GM-CSF. The detectable cytokine concentrations at two time points in the menstrual cycle are summarized in Table 4. No statistically significant differences between cycle days 9 and 20 were found. Although GM-CSF and IL-5 were analyzed, no detectable levels were found.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

Our laboratory previously described a technique for the collection of mucosal secretions (3). Initial studies suggested that the extraction procedure used to recover Ig from the sponges was not appropriate for cytokine recovery. IL-4, IL-2, and IFN- were not obtained from the sponges and IL-10 and IL-6 were extracted at very low rates using the method previously described. In order to quantitate antibody or cytokine concentrations, an efficient and consistent processing step was necessary to ensure release of protein from the sponges.

Cytokine recovery was increased by altering the existing protocol. The final extraction procedure utilized the standard extraction buffer previously described, but the sponges were incubated at 4°C for 30 min with extraction buffer prior to the first 15-min centrifugation at 16,000 × g. A second wash with 300 µl of extraction buffer followed by immediate centrifugation was also incorporated into the final protocol. These changes in the extraction procedure resulted in significant increases in the amount of cytokine obtained without altering the rate of Ig recovery. Of the nine cytokines examined, only IFN- and IL-4 bound within the sponges and were essentially impossible to extract. More than 50% each of the IL-6, IL-10, and GM-CSF loaded on the sponges was recovered with the new technique. IL-1, IL-2, IL-5, and IL-12 had recovery rates greater than 85%. Consistency in the ability to recover cytokines was demonstrated through repeated experiments; variation between extractions was only 3 to 7%. Although 100% of the individual cytokines are not obtained from the sponges, investigators can be confident that this method can extract cytokines from the sponge matrix in a reproducible and reliable manner.

The depressed recovery may be attributed to a number of factors. It may be due to physical entrapment of the cytokine within the sponge matrix or to chemical degradation of the cytokines. In particular, IFN- and IL-4 have inherent stability problems, suggesting that the molecules are unstable in the extraction buffer, or have altered conformation such that the quantitation by ELISA methods is inaccurate. An alteration in the structure of IL-4 or IFN- may also expose residues in the molecule that react with the polymer matrix of the sponge, resulting in binding and entrapment of cytokine within the sponge matrix (13). The only structural observation that corresponded to recovery rate involved the number of disulfide bonds present in the cytokine structure. IL-2, IL-5, and IL-12 each contain only one disulfide bond (2). Their recovery rates were 85, 90, and 86%, respectively. Cytokines with two disulfide bonds include IL-6, IL-8, IL-10, and GM-CSF (2), and the recovery rates for these cytokines were 50, 59, 55, and 63%, respectively. IL-4, with three disulfide bonds, had only a 5% recovery rate. This trend suggests that percent recovery decreases with increases in the number of disulfide bonds. However, the trend did not hold true for IFN-, which has no disulfide bonds. Why IFN- is lost on the sponges is unclear.

The ophthalmic-sponge technique was used to analyze cytokine concentrations in the cervical secretions of women using oral contraceptive pills. Of all the cytokines analyzed, IL-6, IL-8, IL-10, and IL-12 demonstrated consistent detectable levels in the secretions. This result differs from those of previous studies examining cervical mucus (5), where only low levels of IL-6 and IL-10 were demonstrated and the cytokines in samples from most subjects were below the level of assay detection. This difference may be accounted for by the different collection method and ELISA kits used for the concentration determinations. IL-8 concentrations in users of oral contraceptive pills were higher than concentrations of other cytokines both early and late in an oral-contraceptive-mediated menstrual cycle (138.3 ± 52.5 and 54.4 ± 11.2 ng/ml, respectively). No statistically significant differences were noted in the concentrations measured at day 9 or 20 of the menstrual cycle for any of the observed cytokines. This is unlike IL-8 or IL-6 concentrations in healthy cycling women, who demonstrate significant variability in the concentration of each cytokine over the menstrual cycle (4, 5). Similar findings were obtained by using other collection methods and for ovulating women (1, 10). The measurement of various cytokines in the cervical secretions demonstrates the complexity of the microenvironmental milieu of the cervix. These data represent an assessment of cytokine levels in healthy women and can be used to compare changes induced by infection or by the use of intravaginal products.

In summary, the ophthalmic-sponge technique is a useful and simple method for isolation of mucosal secretions. Consistent recovery of both Ig and cytokines allows for analysis of secretions from various mucosal sites. A limitation of this technology is the fact that at least two tested cytokines, IL-4 and IFN-, are not recoverable from the sponges. This suggests that investigators should test for binding to the Weck-Cel prior to initiating studies of a new cytokine or protein. However, the impact from data generated using this technique in vaccine and drug trials will significantly increase our knowledge of mucosal immunity and should far outweigh the limitations of the method.