In a previous work [21], we have shown that CBDs affect the technical properties of paper fibres (secondary fibres from the paper mill Portucel).
The concentration of CBDs, used in those experiments, was in the range
of 1–2 mg of CBD per gram of fibres. It is arguable whether this
relatively low amount of protein is sufficient to cause modifications
in the fibres' interfacial properties. This would probably imply a
substantial coating of the fibres by CBDs. In this work, we analyzed
fibres from Portucel treated with CBDs conjugated to FITC,
and attempted to estimate the percentage of surface coverage. Fibres
treated with only CBDs didn't present any fluorescence. As may be
observed in Figure 1,
the fibres do not display a uniform distribution of fluorescence. This
may be due to the chemical heterogeneity (lignin/hemicellulose) and/or
to the variable crystallinity [24].
The regions with less intense fluorescence in the picture were selected
for CBD-FITC quantification, since these regions are expected to be
more crystalline (Pinto et al., unpublished). The detected
fluorescence is produced by CBDs adsorbed on both sides of the fibres
(top and bottom). Indeed, the fluorescent radiation crosses the fibres
with just a slight reduction in intensity [25]. Figure 2
shows a Whatman CF11 fibre, both on bright field and fluorescence
microscopy. As it can be observed in the circled area, this cellulose
has a rather smooth surface. The extremities of the fibres are expected
to have a higher number of fissures and loosen microfibrils, increasing
the available area, and consequently the adsorbing sites for CBDs [12], as indicated by the higher fluorescence emission (Fig. 2). The adsorption of CBDs on Whatman CF11 fibres was analyzed, using a protein concentration of 2 mg/gfibres. The estimated surface concentrations of CBDs adsorbed on Portucel and Whatman CF11 fibres (Fig. 1) are shown in Figure 3.
Figure 1. Portucel and Whatman CF11 fibres treated with CBD-FITC. The fibres treated with a concentration of 60 μg/mL (or 2 mgCBD/gfibres).
The images were acquired with an exposure time of 600 ms. The white
squares identify the areas selected for analysis. The more fluorescent
parts (black regions on the analyzed images) are out of range
(calibration shown elsewhere), and were therefore excluded from the
analysis.
Figure 2. Whatman CF11 images. The characteristic curled structure of the fibres (circle).
Figure 3. Estimated surface concentration of adsorbed CBD.
The estimated fraction of surface coverage is indicated in the figure
bars. The values shown are based on the assumption that the protein is
adsorbed on the external surface of the fibres only. The values are
shown with 95% confidence intervals error bars.
In another experiment, Whatman CF11, amorphous cellulose
and Sigmacell 20 fibres were allowed to adsorb CBDs from a much more
concentrated CBD solution (400 μg/mL), corresponding to 20 mg/gFibres (Fig. 4). This concentration is expected to saturate the fibres according to the adsorption isotherm.
Figure 4. Images obtained by fluorescence microscopy of Whatman CF11, amorphous and Sigmacell 20 fibres.
The images were obtained before – 1st and 3rd column – and after – 2nd
and 4th column – the image analysis of fibres treated with a CBD-FITC
concentration of 400 μg/mL. The black
areas on the treated images correspond to the fluorescence emission,
which is higher than the one used in the calibration. Fluorescence
images acquired for 100 ms (CF11) and 80 ms (Amorphous and Sigmacell).
Considering the size [26] of a cellobiohydrolase I CBD (3.0 × 1.8 nm), the density of a CBD monolayer corresponds to 3.08 × 10-13 mol/mm2.
It is also considered that the fluorescent signal is produced by CBDs
adsorbed on the two sides of the fibres' external surface. Indeed, the
accessible area may be much larger than the one corresponding to a
flat, impenetrable fibre. Therefore, an estimate of about 4 layers of
CBDs adsorbed in the Whatman CF11 surface results from this reasoning,
when the larger concentration of CBDs is used; for a lower CBD
concentration, a surface coverage of 77% and 110% is estimated,
respectively, for Portucel and Whatman fibres (Fig. 3),
dividing the calculated surface concentration by the theoretical
monolayer of CBDs. This is much higher than the expected maximum of one
layer of adsorbed CBD, at saturation. Although the surface is
apparently smooth, the fibres may have irregularities, such as
microfissures or holes created by CBDs [27].
The presence of more or less loose microfibrils, large pores or
fissures may substantially increase the surface area and the amount of
bound proteins, thus leading to a higher fluorescence emission than the
theorized monolayer. Indeed, confocal microscopy reveals that the
surface has many irregularities (Fig. 5-c), while the inner region presents a homogeneous structure (Fig. 5-a).
Another important observation provided by confocal microscopy was that
fluorescence in the inner core of the fibres was always superior to the
background intensity (Fig. 5-d and Fig. 5-e), indicating that CBDs may have penetrated into the fibre. As a matter of fact, as shown in Figure 5-d,
fluorescent material (CBD-FITC) was detected at all depths of the
fibre. This observation is supported by immunolabelling of CBD-treated
CF11 fibres (Fig. 6).
This analysis revealed the presence of CBDs (black spots) in the
interior of the fibre. Considering this, we may now explain that the
surface concentration of CBDs in Figure 3
arises from the CBDs penetration deep inside the fibres. Another aspect
to take into account is that CBDs may not adsorb as a single and well
ordered monolayer, but rather as agglomerates [28],
thereby increasing the average fluorescence per unit area.
Nevertheless, it is quite probable that the CBD coating of the external
surface is rather significant.
Figure 5. Images from confocal microscopy. Three views of a CF11 fibre are shown, as schematized in the right insertion of figure c. The insertions d, e and f correspond
to the pixels intensity (256 grey levels) obtained at the position
indicated by the line (white circle), at different depths. The
adsorption conditions were 20 mgCBD/gFibre, for 30 minutes of contact. Each image corresponds to an acquisition thickness of 1 μm.
Figure 6. Electron microscopy image of immunolabelling of CBD-treated Whatman CF11 fibres.
Amorphous cellulose was prepared, by treating Whatman
CF11 with phosphoric acid, which induces the swelling of fibres. As it
can be seen in Figure 4, the fibres are larger than the original ones (amorphous Whatman versus Whatman
CF11). This swelling effect is due to the disruption or loosening of
the microfibrils, thus increasing the volume occupied by the total
fibre. As a result, the total surface area available for CBD adsorption
increases. Consequently, it is expected that the measured CBD
fluorescence would be higher than the one obtained for CF11 fibres,
mostly due to an easier penetration of CBD into the fibres structure.
This was confirmed, as can be seen in Figure 3.
The surface coverage increases about 50% (4 versus 6 layers,
respectively for Whatman CF11 and amorphous cellulose). This result can
arise either from increased CBD affinity for the more amorphous fibres,
or to easier penetration, and hence higher concentration in the fibres.
Due to the higher fluorescence obtained with these fibres, the majority
of the image exceeded the maximum concentration used in the calibration
and it had to be excluded in the image analysis (Fig. 4).
Sigmacell 20 is obtained by the separation of crushed cellulose fibres, with an average size of 20 μm.
This mechanical treatment expectedly leads to broken ends or loosen
fibrils (amorphous material). Then, the adsorption of CBD is expected
to be higher than with CF11 and comparable to the amorphous cellulose
(see Figure 3). Again, the fibres present a high fluorescence emission corresponding to a high amount of adsorbed CBD: about 5 layers.