The number of penetration glands and their ability to be stained by various dyes have already been shown previously for Trichobilharzia regenti cercariae [5, 11]. In this study, we used some other techniques for fluorescent visualization of acetabular glands which are applicable in confocal microscopy, namely labelling by alizarin and Alexa Fluor® 488.
The ability of alizarin to stain CA gland content in schistosome cercariae has been previously exploited in light microscopy - e.g. . It is capable of marking structures containing calcium; this element has been shown to occur in CA glands of Schistosoma mansoni in a high concentration , most likely in an ionic form . However, it is known that while being a weak complex-forming agent with respect to the free Ca2+ ions, alizarin exhibits affinity to calcium in biological structures, forming heteroligand complexes with biological objects . Similarly to S. mansoni, the presence of a significant amount of calcium in CA glands of T. regenti was proven previously . Thus, we employed alizarin for visualization of CA glands in confocal microscopy as it produces fluorescence upon excitation at 488 nm (see Figure 1D). Cercariae had to be alive because alizarin stains the glands intravitally and starts to leak out of fixed specimen. For this reason we immobilized cercariae by Procain anaesthesia. This was previously used for the purpose of photography of live cercariae in fluorescence microscopy . However, immobilization is neither permanent nor total, which is problematic for z-scanning using a confocal microscope. Therefore, the number of completely scanned cercariae was not high. Nevertheless, we confirmed that the four gland cells described originally as preacetabular occupy in fact the space around the acetabulum; therefore, they should be called rather circumacetabular glands.
During investigations of cercarial musculature  by labelling with FITC-phalloidin for actin filaments we found that acetabular gland cells express autofluorescence. Besides this, various fluorescent probes have been tested to visualize the glands and secretory vesicles - Alexa Fluor® 488 was the most useful one. All the approaches mentioned above were successfully used for 3-D reconstruction of gland shapes or for estimation of gland/cercarial body volumes.
Recently, Collins et al.  published their work on labelling of S. mansoni cercarial glands using fluorescent lectins . Comparing our three-dimensional model of T. regenti cercarial glands with S. mansoni[7, 8, 35] showed that not only the number and types of glands are common features of both schistosomes, but also the organization, shape and gland openings look very similar. The authors  mentioned the previous work of Dorsey and Stirewalt and Dorsey et al. [7, 10] who estimated that S. mansoni glands occupy nearly two thirds of cercarial body volume; unfortunately, none of these three papers brought exact data based on measurements. We believe that their estimation based just on visual observation was very inaccurate as our initial supposition for T. regenti relative gland volumes was also higher than the results of stereometry (one third of the body volume).
We confirmed that there are four visible openings on the apical surface of cercaria from which the contents of acetabular glands are released. Our data enable us to suppose that it is highly unlikely that the number of the openings may vary depending on branching of gland ducts outside bundles, as mentioned by these authors for S. mansoni. Each bundle containing both PA and CA ducts is enveloped by a common muscle layer which supposedly controls release of secretory vesicles. Therefore it is probable that both gland types are released together during penetration into host skin. This is in accord with the hypothesis that the two separated compartments (PA and CA secretory cells) contain substances which may be activated after mixing of their contents outside cercarial body . However, we are aware of in vitro experiments in which sequential emptying of particular penetration gland types was observed . This process certainly deserves further research.
Some doubts have existed about the architecture of the head gland in S. mansoni cercaria. It was pointed out that it might be composed of two cells or at least contain two nuclei , although this statement was rather speculative. During our experiments we have not found any evidence that there could be two cells composing the head gland in T. regenti.
The features of secretory vesicles produced by particular gland types are markedly affected by the mode of material treatment for TEM. From this point of view, high-pressure freezing and freeze substitution method gave the best results and more structural details could be evaluated. In general, it can be stated, that the vesicles in T. regenti glands resemble those in corresponding gland types of S. mansoni (reviewed by Dorsey et al. ). A diference was seen in size of the vesicles which were 2-3 times larger in case of T. regenti. This is probably related to the body size differences between T. regenti (total length ca. 760 μm ) and S. mansoni cercaria, which is smaller (ca. 500 μm; ). Although the composition of acetabular glands of T. regenti seems to be different to some extent from that of S. mansoni (e.g. in terms of types of proteolytic enzymes, presence of mucous glycosubstances [11, 22, 28, 30]), this obviously has a little effect on the appearance of secretory vesicles in TEM. In CA vesicles of T. regenti, similar electron-lucent granules were observed as in S. mansoni, to which calcium ions were localized . This is another (indirect) proof of the presence of calcium in CA glands of T. regenti which may have consequences in interpretation of calcium function in these gland cells. It was confirmed, that the role of Ca2+ could be in regulation of proteolytic (serine peptidase - cercarial elastase) activity within (or outside) CA glands [37, 38]. However, it has been proven by various methods that T. regenti does not possess this enzyme and instead it most likely uses cysteine peptidases (e.g. cathepsin B2) localized in PA glands [11, 28, 30, 39].
The three approaches to estimate volumes of cercarial glands gave similar results. Both fixed and live cercariae were used in these experiments to compare the effect of fixation on the shrinkage of cercarial bodies and of gland cells. The total volumes of fixed cercarial bodies decreased to ca. 69% of that of live anaesthetized cercariae, but the relative gland volumes remained similar in both cases, ie. between 14.4% - 19.5% for PA glands and 12% - 15,2% for CA glands, with low variations depending on the method used. Relatively high standard deviations were caused by natural variability among individual cercariae and probably also by their eminent contractility. In the case of voxel counting method for evaluation of optical sections obtained by confocal microscopy, there may be a little over- estimation due to inappropriate image enhancement processing antecedent to volume estimation. The estimation of head gland relative volume was 6% using TEM and stereology; we did not evaluate its volume by confocal microscopy as we have not succeeded in finding a specific fluorescent marker for this gland. All three gland types together fill around 1/3 of cercarial body which stresses the importance of these secretory cells for the cercarial stage of schistosomes employing their secretions during invasion of vertebrate hosts. These data will help in the future to estimate quantities of bioactive molecules found within the glands of T. regenti. Unfortunately, no volumetric data exist from S. mansoni model, making exact comparisons impossible. The volumes could be just adequate to the size of cercarial body, which is about 1.5 × longer than that of S. mansoni cercaria [5, 10]. But, supposedly, the relatively high volume of gland contents enables the cercariae of bird schistosomes to penetrate through the tough skin on exposed duck legs. Also, it may be the cause of higher invasion efficiency by Trichobilharzia compared to Schistosoma mansoni in the case of human skin .
The estimation of pH inside acetabular glands of T. regenti was quite a difficult task. Although fluorescent probes like SNARF-1 expressing shift in the wavelength of emitted light depending on pH are commonly used for pH estimations in cells or their compartments (e.g. ), in our experiments it was necessary for this compound to cross the cercarial glycocalyx, syncytial neodermis including its basal lamina and cell membranes of gland cells. For the purpose of calibration of the system by a set of buffers of known pH, nigericin was used as H+ ionophore to disrupt pH gradients. This might to some extent influence the real picture about the actual pH within the glands. However, at least a tendency could be seen of higher pH in CA comparing to PA glands. This result was anticipated as CA cells contain a relatively high amount of Ca2+ cations - their anionic counterpart is not exactly known but it could be hydrogen carbonate and carbonate in S. mansoni according to . Surprisingly, we recorded higher pH in PA cells than expected. It was shown that PA glands produce a cysteine peptidase (cathepsin B2) which has an optimum of activity at pH 6.0 ; its activity drops dramatically at pH above 6.5. Therefore it can be hypothesized, that the higher value of pH in PA cells (around neutral) could restrain the hydrolytic activity of cathepsin B2 in order to prevent autolysis of cercaria. The activity of this acid-active peptidase should be restored upon the contact with the stratum corneum of the skin, which has pH usually between 4 - 6.5 in humans  and 5.4 - 6 in domestic ducks  due to the presence of organic acids forming an antimicrobial protective barrier ("acid mantle") in the upper layer of vertebrate skin.