Background
Cervical mucus is reported to regulate sperm penetration and transport to the upper reproductive tract [
1,
2]. It also provides lubrication to the cervix by enhancing its wetness and thus preventing its desiccation, and retards enzymatic degradation of the cervix and providing it with protection from pathogenic invasion and infection [
3‐
5]. Its secretion, at a rate of 20–60 mg per day acts as a fence to sperm and pathogen entrance [
6]. Although a reduction in mucus viscosity may allow foreign agent penetration, millions of micro-organisms a day are reported to be cleared from the reproductive tract by cervical secretions that are the tract's most effective first line of defence [
7].
Thus far six mucin genes have been reported to be expressed by the female reproductive tract, namely MUC1, MUC2, MUC4, MUC5AC, MUC5B and MUC6 [
6]. The genes for MUC2, MUC5B, MUC5AC and MUC6, are found on chromosome 11p15.5 and express the secreted gel forming mucins, whereas MUC1 and MUC4 are membrane associated mucins expressed by the epithelium of the ecto-cervix and vagina [
7]. Of these, MUC4 and MUC5B are reported to be the major mucin genes expressed by the endo-cervix [
8]. The variation, under hormonal influence, of the viscoelastic and rheological properties of these mucins during the menstrual cycle is well documented [
4].
Human crude saliva is known to inhibit Human Immunodeficiency Virus type 1 (HIV-1) activity in an
in vitro assay [
9,
10]. These authors speculated that it was the mucus component that inhibited the virus. We very recently showed that both crude saliva and its purified mucin components MUC5B and MUC7 inhibited HIV-1 activity [
11] and so did the purified MUC1 of breast milk [
12]. The MUC1 of breast milk also showed anti-pox viral activity [
13]. Our hypothesis is that cervical mucins should have a similarly inhibitory effect on HIV-1 activity, an important question considering that the vagina and cervix are significant routes for HIV transmission. The aim of this study therefore was to extract and purify the mucins in the pregnancy plug mucus and to determine their anti-HIV-1 activity using an HIV inhibition assay.
We therefore extracted and purified mucins from the pregnancy plug mucus which occludes the cervical canal throughout the pregnancy period [
2,
14]. This large mucus plug which is more like the mucus of the luteal phase than the mucus of the mid-cycle [
2] was obtained during labour and just prior to delivery.
Sub-Saharan Africa is reported to be home to about 25 million adults and children who are HIV positive [
15]. In Southern Africa 25.7% of the population has HIV/AIDS, making this the most highly prevalent region of infection compared to the Eastern and the Western regions with 11.4% and 4.3% prevalence respectively [
16]. In South Africa alone, between 4.68 and 7.03 million people were living with HIV/AIDS in 2004 [
17], of whom 55% were female [
18]. Thus this preliminary study could make a significant contribution to the efforts being made in controlling this epidemic.
In this study we report the anti-HIV-1 activities of crude and purified human pregnancy plug mucus and mucins in an in vitro inhibition assay. We have demonstrated that the purified mucins from the pregnancy plug mucus inhibited HIV-1 infection of the CEM SS cells. However, the crude pregnancy plug mucus failed to inhibit HIV-1 infection of these cells.
Discussion
According to various studies [
9,
10,
21,
22], salivary macromolecules (possibly mucins) aggregate HIV-1 prior to host cell entry, thus preventing transmission of HIV-1 through saliva. Wiggins
et al. [
7] reported that mucus is the first line of defence against pathogenic micro-organisms. Studies in our laboratory have also confirmed these findings [
11]. Crude saliva (from individuals with a self-declared risk free lifestyle and thus presumably uninfected), and its purified mucins MUC5B and MUC7 [
11] and purified MUC1 from breast milk [
12] show anti-HIV-1 activity in an
in vitro inhibition assay.
It thus remains to be asked why other areas such as the female reproductive tract and breast milk, so rich in mucus and mucins quite similar in substance and conformation to those in saliva, still remain major routes of transmission of the virus. In the case of breast milk we showed that its MUC1 component inhibited the HIV-1 from infecting CEM SS cells in an
in vitro assay only after it was dissociated from the milk fat globules and isolated and purified by caesium chloride density gradient ultra-centrifugation. Crude breast milk had no such inhibitory effect on HIV-1 [
12]. In the light of this we decided to investigate whether cervical mucus and mucin display any anti-HIV-1 properties, considering that the cervix is a significant route of transmission in women.
The quality and quantity of cervical mucins during the different phases of the menstrual cycle are reported to vary either through the influence of oestrogen (proliferative phase) or of progesterone (luteal phase). For example the production of MUC5B was reported to increase at the mid-cycle and decrease during the secretory phase of the menstrual cycle whilst MUC4 increases during the luteal phase of the menstrual cycle [
8,
23]. These cyclical variations together with the fact that cervical scrapings, which yielded very small amounts of crude material made it difficult to investigate the anti-HIV-1 activity of these mucins per se. Therefore mucus plugs at the mouth of the cervix rich in mucin [
2,
14], were obtained from women in labour. However, a comparison of the effect of purified plug mucin versus purified cervical mucin on HIV is being planned.
In this study we have demonstrated that the purified mucins from the pregnancy plug inhibited HIV-1 infection of the CEM SS cells. However, the crude pregnancy plug mucus and the media failed to inhibit HIV-1 infection of these cells. Though the mechanism of inhibition is not clear, it is likely that when the HIV-1 was incubated with the mucins, the virus was trapped by aggregation through the sugar side-chains of the mucins, a purely physical phenomenon [
10,
24‐
26], resulting in preventing the virus from entering the host cells (CEM SS cells). This was supported by our finding that salivary MUC7 inhibited HIV-1 infection of the CEM SS cells when it was incubated with the virus prior to addition to the CEM SS cells. However, the mucin failed to inhibit viral infection of these cells when it was incubated with CEM SS cells prior to addition of the virus (unpublished data). This suggests that the mucin inhibits HIV-1 infection by physically aggregating the virus than by blocking putative viral binding sites or receptors on the cells.
The virus and mucins were incubated together with the cells for different incubation periods, i.e. 30 min, 1 h and 3 h to determine the effect of time on infection or lack thereof. Cultures were then washed three times after each incubation period to remove free virus and cultured for another 4 days in IL-2 rich media. This was done to determine if the virus had entered the cells during the initial incubation step and was able to replicate inside the cells for the extended incubation period to produce p24 antigen, or if the mucins were successful in preventing viral entry into the cells and therefore prevent the production of p24 antigen.
To further confirm the hypothesis that mucins inhibit HIV-1 activity by physically aggregating the virus, the CEM SS cells were incubated with the filtrates from the mixtures. The lower infection (2.5%) of the CEM-SS cells by the filtrate from the mixture of HIV-1 plus purified pregnancy plug mucins suggests the presence of insignificant amount of viruses in the filtrate or almost complete aggregation of the virus by the mucins, leaving no free viruses to pass through the filter paper into the filtrate to cause viral infection. On the other hand the 100% infection of the CEM-SS cells caused by the filtrates from the mixtures of HIV-1 plus crude pregnancy plug mucus and HIV-1 plus media suggests the presence of higher amount of viruses in the filtrates or the failure of the crude pregnancy plug mucus and the media to aggregate the viruses. This finding agreed with the report that HIV-1 may bind to the high-molecular weight components which results in macromolecular complex formation which is removable by filtration through 0.45 μm pore filter paper [
10,
24‐
26].
The lack of inhibition by crude pregnancy plug mucus compared to the inhibition by purified pregnancy plug mucins is not clear. However it should be considered that mucins constitute only about 0.5–1% of total crude mucus [
27] which is known to contain water, glycoproteins, lipids, nucleic acids, lactoferrin, lysozyme, immunoglobulins and ions [
7]. It is likely therefore that the potency of mucins would in this case be in their purified form rather than when they are a minor part of a larger secretion in which their concentration would be diluted. This was quite different in the case of crude saliva, the inhibitory effect of which was similar to that of its purified mucins, separable by gel filtration and individually effective against the virus [
11]. However, quantification of the amount of mucins in the crude mucus prior to any assay should be considered before drawing this conclusion.
The heat inactivated HIV-1 (negative control) caused an approximately 30% infection of the CEM SS cells suggesting that the viruses, when inactivated but not completely killed are still infective, albeit to a lesser degree. To determine whether there is a dose/effect relationship and the lowest possible effective concentration with anti-HIV-1 activity, ten fold serial dilutions (10-1to 10-4) of the mucins were also done from a starting concentration of purified mucin of 0.9 mg. The mucins showed strong anti-HIV-1 activity down to a dilution of 10-4, but in this study the lowest possible concentration which can cause inhibition of HIV-1 activity was not identified. Thus a lower starting concentration of purified mucin than 0.9 mg would be advisable.
There was also no effect of time (incubation period) on the inhibitory effect of mucins or the infectivity of the virus. This suggested that the mucins aggregated the virus immediately and permanently. However, shorter starting times of incubation of mucins and the virus would be necessary to determine the shortest time mucins take to aggregate the virus.
Although HIV-1 Subtype C is currently the most prevalent in South Africa, the Subtype D which was used in this study was found during the early HIV epidemic in the country and is quite prevalent here, albeit to a lesser degree. Even though we wished to use the Subtype C strain, the Subtype D strain is unfortunately the only lab adapted strain we had available to us in the vicinity of Cape Town and it is possible that this is the only laboratory based HIV assay in the country. As described in the Methods section, this virus was first isolated from an AIDS patient by the Department of Medical Virology, Tygerberg Hospital, University of Stellenbosch, South Africa, in February 1988, and it was fully characterised and sequenced subsequently [
28]. The human T lymphoblastoid cell line (CEM SS cells), which was used in this study, is reported to express CD4, CXCR4, ICAM-3 and MHC class II molecules [
29]. These cells are capable of developing easily quantifiable syncytia formation in four to six days upon the addition of HIV-1 [
30]. Although Subtype C predominantly uses CCR5, several instances of co-receptor switch to CXCR4 or even dual tropism have been observed in Subtype C, especially later in infection. Therefore this study could be relevant to
in vivo situations, where transmitted viruses are most often CCR5 tropic.
Extraction of mucus was in 6 M GuHCl and proteolytic inhibitors which included 10 mM EDTA, 5 mM NEM, and 1 mM PMSF to reduce endogenous proteolysis of mucins [
2]. PMSF and EDTA inhibit serine and metallo-protease activity respectively whilst NEM inhibits thiol proteases and minimizes thiol-disulfide exchange [
1].
Caesium chloride density gradient purification removes all contaminants such as non-mucin proteins, lipids, proteoglycans and nucleic acids from mucins [
31]. Purification of the mucins was confirmed by SDS-PAGE [
32]. The removal of these contaminants from mucins was believed to be by dissociative conditions through the presence of GuHCl [
1], known to be a widely used denaturant [
33] which in this case could well dismantle the tertiary structure of mucins [
14].
The presence of MUC1, MUC2, MUC5AC and MUC5B in the pregnancy plug mucus was confirmed by Western blotting with MUC2 expressed as a doublet and in small amount compared to the other mucins. Immunohistochemistry confirmed previous reports of the expression of MUC4 and MUC6 by the endometrial tissue (data not shown), but their presence in the mucus plug could not be confirmed due to the lack of antibodies to these mucins for Western blotting. This result agreed with that of Gipson
et al. [
6], Wiggins
et al. [
7], Gipson
et al. [
23] and Wickstrom
et al. [
34], studies which reported the expression of MUC1, MUC2, MUC4, MUC5AC, MUC5B and MUC6 by the female reproductive tract.
Competing interests
The authors declare that have no competing interests
Authors' contributions
HHH carried out the biochemical studies and drafted the manuscript. CdB established and carried out the HIV inhibition assay. ZEL and MGT participated in the biochemical studies. LS participated in pregnancy plug mucus collection and analysis. DK contributed ideas to the design and coordination of the study. ASM conceived of the study, participated in its design and coordination and finalised the manuscript. All authors read and approved the final manuscript.