Cervical mucus is produced by the biosynthetic activity of the secretory cells of the cervix and contains three important components: (1) mucus molecules; (2) water, (3) chemical and biochemical compounds (sodium chloride, protein chains, enzymes, etc.).

Mucus molecules have two important properties: (a) they are able to join together to make polymers or an extended three-dimensional network (i.e. a gel); (b) since they are glycoproteins their properties can vary widely. Thus different types of mucus are produced, for example G, L, S, P and F mucus, which form different networks or gels. Also, other substances--ions, protein chains, and enzymes--are able to modify the interaction of the mucus molecules and, as a consequence, their biophysical properties. The molecular weight of mucus is about 70,000 daltons, and it is believed to be several million daltons for gels. A general structure of a gel is given in Figure 2.

Because of these structures mucus is not a normal but an abnormal (i.e. non-Newtonian) fluid (Löfdahl and Odeblad 1980) and its viscosity is not able to be measured using liquid-flow techniques. It is therefore necessary to use other methods, preferably nuclear magnetic resonance (NMR) techniques which do not involve flow, but make use of thermal movements of molecules in the fluid (James 1975).

As well as the glycoproteins which are secreted, the cells produce membranous glycoproteins which are bound to the cell membrane. These glycoproteins enable immunological identification of the cells. A similar product is the substance located in the basal membrane which is probably secreted by the cells of the mucous membrane, as are also the cell adhesion molecules.

There are three groups of cells in the mucous membrane of the cervix: (1) cylindrical (columnar) secretory cells (the majority); (2) cylindrical (columnar) ciliated cells; (3) "reserve" cells. The origin of the secretory cells is known but the mode of development of the two other groups of cells has not yet been decided. The cells of the mucous membrane are slowly detached and are displaced with the mucus. New cells are formed to replace them.

The biosynthesis of mucus is a complicated process. The epithelial cells are stimulated by oestrogens (S-, L- or P-cells) or by progesterone (G-cells). The hormone is bound to a receptor in the cytoplasm, and is then transported to the cell nucleus. The receptor + hormone complex then activates certain parts of the genetic material (DNA) which is transcribed to another type of genetic material (RNA) which, in turn, carries the genetic message to the place in the cell where the amino acids are arranged in the correct sequence to form the protein core of the molecule. Carbohydrate molecules are then attached by enzymes onto the protein core. The instruction to the cell how this should be done is probably different in cells of type S, L, P and G by information laid down in cells already during their embryonic development (see below).

The response of the cells to oestrogen or progesterone stimuli is comparatively slow (from one to several hours). The S and P cells also seem to have access to a much faster response mechanism, the stimulation by noradrenaline acting on a beta receptor localized in a cell membrane. This response occurs probably within a few minutes and may be responsible for the "instantaneous" discharge that some women experience on acute stress, e.g. in "stage fright" or a sudden emotional upset.

After the preceding presentation of some basic information, I shall now turn to the discovery of the various types of mucus.

Figure 2. Molecular structure of mucus. I, ionic bond;P, peptide bond;H, hydrogen bonding;D, disulfide linkage; V, van der Waals bonding; E, entanglement of mucus molecules.

My research on the cervix began in 1949 during the course of my microbiological studies, which were concerned with mycoplasms in the genital tract of healthy, and sick women. I was responsible for the gynaecological examinations and the collection of microbiological specimens. This research was published in 1951 and 1952 (cf. M‚len and Odeblad 1951, 1952). We made some interesting observations in the healthy women examined during the pre-ovulatory and post-ovulatory phases:

(1) Mycoplasms were cultivated several times in 5 out of 32 non-pregnant married women but in not one of 13 virgin women or in 11 additional virgin women studied by Frisk et al. (1952) (Chi-squared test, P < 0.05). This result was in accordance with a sexual transmission of mycoplasms, as proposed by Dienes et al. (1948).

(2) In the healthy married women 6 out of 17 cultures were positive in the first part but none were positive in the last part of the cycle (Chi-squared test, P < 0.01), an indication that post-ovulatory mucus was able to exercise an antimicrobiological effect (Barton and Wiesner 1945; Pommerenke 1946). 1 shall return to this problem later.

During the collection of samples I witnessed the characteristic variations in mucus in the course of the cycle. I also happened to read three important papers. The first was a review by Esselbom (1947) on cyclical variations in cervical secretions, the second a paper by Rydberg (1948) on crystallization of cervical mucus and the third a paper by Bloembergen et al. (1948) on the new NMR method for measuring viscosity.

The Existence of Different Types of Crypts and of Mucus

I have already mentioned the cyclical variation of the cultures of the mycoplasms. We observed another interesting situation: in three married, healthy, non-pregnant women mycoplasms were recovered in the same cyclical manner despite the fact that it was not possible that these women had become re-infected. If all crypts produced an antimicrobial mucus in the post-ovulatory phases of the cycle all the mycoplasms would be killed and hence not able to be recovered. Then, in 1952, 1 suspected that the mycoplasms could survive in secretory inactive crypts after ovulation.

To investigate the problems associated with cervical crypts I studied biophysics at the University of California, specializing in NMR and radioactivation techniques. After my return to Sweden I commenced in 1954 to apply these methods to the study of the secretions of cervical crypts. In what follows, only the application of NMR is described. Radioactivation studies have supported the conclusions obtained with NMR.

In principle, two approaches were possible:

(1) Investigations on intracanalicular mucus - macrosamples (Figure 3) which were possibly a mixture of two or several types (Odeblad and Bryhn 1957).

(2) Investigations on mucus obtained from within crypts - microsamples (Figure 4) (Odeblad 1966b).

The investigations of the macrosamples were comparatively easy but those on the niicrosamples required new, extremely precise equipment which required several years to develop. During these years I found in the medical literature some support for the hypothesis for the existence of different crypts. Observations by Roland (1958) on crystallization and by Montgomery (1959), on the composition of mucus stimulated my research and in 1959 I presented for the first time the results of microscopical examinations which showed that cervical mucus was composed of several different types which are produced by different crypts (Odeblad 1959).

In 1966 (Odeblad 1966b) I succeeded in proving, by examination of microsamples, the existence of crypts ("glands") which responded differently to the same hormonal stimulation. Among 70 crypts studied NMR recordings and slide samples indicated that 38 crypts contained only one type of mucus with low or high viscosity, and 11 crypts probably contained a mixture of mucus with both high and low viscosity. 21 microsamples were contaminated during the extraction procedure. The existence of crypts which contained two types of mucus was later proved and published by my collaborator (Rudolfsson 1971). Probably these crypts have two branches with a common opening (Figure 1), the branches having different secretory functions.

Figure 3. Microsample of mucus in the cervical canal.

Figure 4. Microsample of mucus in a cervical crypt.