1.

Introduction

The genus Fusarium is a well-known plant pathogen that causes damage to agriculture.^1,2 Over the past 30 years, the Fusarium species have been intensively studied and of the 61 identified species, at least 35 are known to produce mycotoxins that are harmful to animals and humans.^3–8

The Fusarium mycotoxins are known as chemically different secondary metabolites. These mycotoxins can contaminate cereal grains causing disease and death in the animals that consume them. The trichothecenes [T2 toxin, deoxynivalenol, nivalenol, fusarenon-X (FX), and diacetoxyscirpenol (DAS)], fumonisins, zearalenone, and moniliformin are the most important Fusarium mycotoxins.^9,10 These mycotoxins cause neurological damage and are frequently associated with animal and human diseases.^4,5 Several mycotoxins are stable under normal food-processing conditions; thus, they may be present in natural foods and processed products.^11,12

Individuals who consume contaminated food are not exposed to a single mycotoxin; rather, they are exposed to a cocktail of them.¹³ In vitro experiments have shown that a mixture of mycotoxins results in additive toxicity or synergy that reduces the effective toxic dose of each component in the mixture.^14,15 If this potentiating occurs in vivo, then the health consequences will be more severe.

In humans, the species Fusarium may cause local, invasive, or disseminated infections. The pathogen typically affects immunocompromised individuals because infections in immunocompetent individuals are reported less frequently.^16,17 Among immunocompromised patients, Fusarium is the second most common cause of fungal infection, with an increased incidence in patients with hematologic malignancies and especially in those exposed to recipients of hematopoietic stem cell transplantations.¹⁸

The entry points to the disseminated infection include the respiratory tract, gastrointestinal tract, and skin.¹⁶ Skin lesions, which are typically located on the extremities, are often described as painful subcutaneous nodules that may ulcerate and become necrotic lesions of the gangrenous erythema type.¹⁸ For skin infections, the entry points are wounds, digital ulcers, onychomycosis, and paronychia.^17,19–21

In recent years, the number of Fusarium onychomycosis cases in immunocompetent patients has increased significantly.^17,22 Guilhermetti et al.¹⁷ reported a 7.5% incidence of confirmed cases of onychomycosis by Fusarium spp. and Fusarium oxysporum (F. oxysporum). The latter is one of the most common and pathogenic of the genus Fusarium and is the specie that has been mostly isolated.^18,22

In humans, this specie is an important emerging pathogen due to the increasing number of severe cases reported and its wide-ranging resistance to antifungal drugs.^23,24 The increase in F. oxysporum infections in immunocompetent patients is an important aspect; thus, the evaluation of the biological and physicochemical effects of F. oxysporum on the skin of healthy rats and their relation with induced apoptosis is of utmost importance.

It is well recognized that techniques based on infrared radiation provide an important route to access chemical bonding information of biological tissues.^25–30 This can be done by exploring the finger print characteristics of the optical absorption bands in this spectral region. Among these methods, Fourier transform infrared photoacoustic spectroscopy (FTIR-PAS) has the special characteristic of allowing depth profile studies to detect the penetration and interaction of substances in biological tissues.^31–33 It can be applied for measurements at clinical conditions because minimal sample preparation is needed. It is a nondestructive method and permits inspection in opaque and highly scattering samples.

Therefore, the aim of this study was to investigate the biological and physicochemical effects of a crude extract (CE) of F. oxysporum on the skin of healthy rats. The FTIR-PAS spectroscopy was applied to evaluate the induced physicochemical structural changes and their relation with the occurrence of apoptosis.

2.

Materials and Methods

The study protocol was reviewed and approved by the Local Ethics Committee for Animal Experimentation of the Universidade Estadual de Maringá (UEM), under protocol number 005/2010 and statement 080/2010.

2.3.

Physicochemical Analysis of the Skin by FTIR-PAS

For the physicochemical analysis of the skin following topical application of either the CE or the CZ medium, two animals/time were used for each experimental group. After each animal sacrifice, the skin containing both epidermis and dermis was removed and sent immediately for ex vivo analysis using an FTIR spectrometer (Varian, Model 7000) equipped with an MTEC model 300 photoacoustic detection cell. Each skin sample was positioned inside the cell chamber with its epidermal side upside down allowing the radiation to excite the dermal region, the opposite side in relation to where the CE or the CZ medium was applied, as illustrated in Fig. 1. After that the cell chamber was purged with the helium gas for approximately 5 min to remove air moisture, and subsequently it was sealed to perform the measurements. This procedure increases the sensitivity of the technique and minimizes spectral interference from loosely bonded molecules of the air such as ${\mathrm{CO}}_2$. All the spectra were collected in the rapid scan mode in the spectral range between 4000 and ${400\mathrm{cm}}^{-1}$, with ${4\mathrm{cm}}^{-1}$ resolution, an average of 28 scans and the interferometer mirror velocity ($v$) of $0.64\mathrm{cm}/\mathrm{s}$ (10 kHz). The reference spectrum was acquired by the use of a carbon black sample.

Fig. 1

Representation of the skin topical application of the Fusarium oxysporum crude extract (CE) or Czapek–Dox (CZ) culture medium, and the sample positioning in the photoacoustic cell.

The depth analysis in the sample was estimated through the thermal diffusion length (cm) as $\mu ={[D\lambda /(2\pi v)]}^{1/2}$, in which $D$ is the sample thermal diffusivity (${\mathrm{cm}}^2/\mathrm{s}$) and $\lambda$ is the incident radiation wavelength (cm). This parameter is the dimension over which the thermal wave decays to $1/e$ of its original amplitude and can be used in the analysis as approximately the sampling depth where the PAS signal is generated. In our experimental conditions, using the skin thermal diffusivity measured before as ${4.1\times 10}^{-4}{\mathrm{cm}}^2/\mathrm{s}$ (Ref. 34) and $v=0.64\mathrm{cm}/\mathrm{s}$, we can calculate the values of μ according to the incident radiation wavelength, being, e.g., 2.8 μm at ${1250\mathrm{cm}}^{-1}$ and 1.7 μm at ${3850\mathrm{cm}}^{-1}$. Therefore, since the exciting radiation was incident on the dermal side of the skin and the application of the CE and CZ medium was done on the epidermal side, the detection of optical bands would be an indication that the formulation penetrated throughout the skin.

3.

Results and Discussion

The obtained CE is a complex mixture produced by fungus growth, which includes metabolism products such as secondary metabolites like mycotoxins. The extract also contains partially digested culture medium residue components. The spectral analyses of organic compounds by FTIR represent a method for obtaining a “fingerprint” of the molecules, i.e., allowing for their identification from the recognition of the functional groups to which they belong. The bands between 1700 and ${1500\mathrm{cm}}^{-1}$ have been related to amides I and II, and they are frequently used as an indicator of secondary protein structure.^35,36 The bands between 3000 and ${2830\mathrm{cm}}^{-1}$ are related to the symmetric and asymmetric stretching of $─\mathrm{C}─\mathrm{H}$ groups in proteins and lipids^36,37 and those between 3800 and ${3000\mathrm{cm}}^{-1}$ to the ${\mathrm{OH}}^{-}$ and amine ($\mathrm{N}─\mathrm{H}$) molecules.³⁶ It is known that the identification of amides I and II and $─\mathrm{C}─\mathrm{H}$ groups may allow the monitoring of conformational changes detected by the vibrational variations of these molecules when subjected to a state different from that considered in physiological tests.³⁸

Figures 2(a) and 2(b) show the FTIR-PAS absorption spectra between 1900 and ${1250\mathrm{cm}}^{-1}$ of the dermis of rats without the application of any substance and at 3, 6, 12, and 24 h after the topical application of both CZ culture medium or F. oxysporum CE. The changes in the absorption bands suggest structural alterations of amide I (1720 to ${1580\mathrm{cm}}^{-1}$), amide II (1580 to ${1470\mathrm{cm}}^{-1}$) and $─\mathrm{C}─\mathrm{H}$ (1470 to ${1350\mathrm{cm}}^{-1}$) groups, which occurred in the dermis after the topical application of the F. oxysporum CE. These spectral modifications may imply that the dermis underwent physicochemical changes that were not observed in the untreated normal skin or the skin on which the culture medium was applied. The region around 1900 to ${1700\mathrm{cm}}^{-1}$ was more intense for the 3 and 24 h control samples. One possible explanation for this difference may be the presence of higher amounts of adipose tissues in the illuminated areas. We should stress that this spectral region was not considered in our interpretation. Moreover, Fig. 3 shows the FTIR data in the spectral region between 3850 and ${3000\mathrm{cm}}^{-1}$. It is observed that the ${\mathrm{OH}}^{-}$ and amine ($\mathrm{N}─\mathrm{H}$) bands presented an increase in their intensities with time after the application of both CE and CZ medium. Once these bands are present in both, this result confirms that the CE and CZ medium permeate through the skin reaching the dermal region. Therefore, our hypothesis is that the physicochemical changes in the skin dermal side observed from the FTIR-PAS data strongly suggest that either partial or complete extract permeation occurred and then produced the observed skin structural alterations.

Fig. 2

Spectra of the dermis of rats obtained by Fourier transform infrared photoacoustic spectroscopy (FTIR-PAS) showing the regions corresponding to the amide I (1720 to ${1580\mathrm{cm}}^{-1}$), amide II (1580 to ${1470\mathrm{cm}}^{-1}$), and ${─\mathrm{CH}}_3$ group (1470 to ${1350\mathrm{cm}}^{-1}$), performed at 3, 6, 12, and 24 h after the topical application of the F. oxysporum CE (a) and after the topical application of the CZ culture medium (b). The dermis without any substance applications is also shown.

Fig. 3

FTIR-PAS spectra at the range between 2500 and $3750{\mathrm{cm}}^{-1}$ of the dermis control, dermis after the topical application of CZ culture medium and dermis at 3, 6, and 24 h after the topical application of the F. oxysporum CE.

Figure 4 shows the behavior of the peak area at approximately ${1370\mathrm{cm}}^{-1}$ within the ${─\mathrm{CH}}_3$ group region (1350 to ${1480\mathrm{cm}}^{-1}$) for 3, 6, 12, and 24 h after the topical application of the F. oxysporum CE. The control point ($t=0$) represents the peak area of the dermis without the application of any substance. In the infrared region, the ${─\mathrm{CH}}_3$ group also gives rise to absorption in the ${1460\mathrm{cm}}^{-1}$ region. At this frequency, the molecule has an asymmetric bending vibration mode. This group also has vibrational characteristics at ${1374\mathrm{cm}}^{-1}$, which is a symmetric bending vibration mode.³⁶ The spectra obtained for the epidermis and dermis after the application of the CZ medium showed different behavior from that observed after the application of the CE, suggesting that the medium and extract have different contributions to the potential structural changes of the highlighted groups (amide I, amide II, and ${─\mathrm{CH}}_3$) as evidenced in Fig. 2(b).

Fig. 4

Behavior of the peak area at approximately ${1370\mathrm{cm}}^{-1}$ within the ${─\mathrm{CH}}_3$ group region (1480 to ${1350\mathrm{cm}}^{-1}$) as a function of time after the application of the CE. The control point ($t=0$) represents the dermis without the application of the CE. The line is a visual guide.

Histopathological evaluation did not show macroscopic changes on the skin surfaces of the rats treated with the F. oxysporum extract. However, during the microscopic examination of the sections stained with H&E, the following changes were detected: (1) red cells were observed on the surface of the skin (at 3 and 6 h), (2) keratinocytes were present on the epidermis with a bright, perinuclear halo that is typical of hydropic alterations, mainly at 3 h, and (3) the cell number increased on the dermis, and the cells were diffusely localized, particularly for the higher frequencies of fibroblasts and macrophages, but few neutrophils and no formation of typical inflammatory infiltrates were observed, as shown in Fig. 5. The inflammation score is shown in Table 1.

Fig. 5

Rat skin photomicrography after the topical application of the CZ medium (a) or F. oxysporum CE (b, c, and d). Note the normal skin morphology 24 h after the medium application in (a). In the skin treated with the CE, the number of dermal cells (b), hydropic alterations (arrows, c), and red cells on the skin surface (arrow, d) increased after 3 h (H&E).

Table 1

Score representing the frequency of skin inflammatory cells 3, 6, 12, and 24 hs after the topical application of the Czapek–Dox (CZ) medium or the F. oxysporum crude extract (CE).

Skin morphometry analysis showed the epidermis of the animals treated with the CE were thicker ($P<0.05$) after 6, 12, and 24 h [Fig. 6(a)]. The dermis was thicker after 12 and 24 h [Fig. 6(b)]. For the animals treated with the CE and death after 3 and 6 h, the thicknesses of the dermis were significantly reduced [Fig. 6(b)]. Bhavanishankar et al.³⁹ suggested that several toxins produced by Fusarium act in different manners when they appear in different combinations. The authors evaluated the variations in epidermis thickness in vivo in guinea pigs after the topical application of isolated toxins or a combination of toxins produced by F. oxysporum. The epidermal thickening by T2 toxin combined with either FXor butenolide (Bd) was greater than by T2 alone. This finding indicates an antagonistic relationship between these toxins. A greater thickness occurred upon the application of a combination of DAS, which is a trichothecene toxin, and Bd, which is a nontrichothecene toxin, suggesting a synergistic toxic effect by this combination.

Fig. 6

Thickness measurements (μm) of the epidermis (a) and dermis (b) 3, 6, 12, and 24 h after the topical application of the CZ medium or F. oxysporum CE. *$P<0.05$ when compared to the control at the same time point, $N=4\text{animals}/\text{time}$.

In the dermis of the animals treated with the CE, the area covered by collagen type I increased after 3, 6, and 12 h, and the area covered by collagen type III increased after 3, 6, 12, and 24 h compared to the dermis of the animals treated with the culture medium [Fig. 7(a)]. For all the evaluated time points, the total area covered by collagen was larger ($P<0.05$) in the groups treated with the CE [Fig. 7(b)]. Figure 7(c) shows the proportion of collagen III:I. The greatest disparity in the ratio of the collagens fibers (6 and 24 h) occurs after the maximum for death (3 h) and proliferation (12 h) of fibroblasts. With the death of cells (fibroblasts), the matrix is reabsorbed. A slight decrease in the mature collagen, type I, is observed in skin treated with the CE. With the formation of new cells, a matrix is produced, and the increase in the immature collagen, type III, is observed. In the control group, a balance in the proportions of the types I and III fibers could be observed.

Fig. 7

The area (${\mu \text{m}}^2$) covered by total collagen (a) and types I and III collagen fibers (b) and the proportion of collagen III:I (c) on the dermis of the rats 3, 6, 12, and 24 h after the topical application of the CZ medium or F. oxysporum CE. *$P<0.05$ compared with the control group, $N=4/\text{time}$.

TUNEL immunohistochemical reaction to the stained apoptotic cells results are shown in Table 2 and Figs. 8(a) and 8(b). The basal layer of the epidermis showed the highest frequency and positive intensity after labeling with TUNEL, followed by the Prickle-cell layer and the other layers that had more diffuse labeling. In the dermis, fibroblasts and hair follicles cells were the most immunostained cell types. The strong staining of the follicles was most likely due to its involutive process that occurred after epilation.⁴⁰ A large number of stained cells were observed in the skin tissue and panniculus carnosus muscle cells. In a previous study, we reported the occurrence of programmed cell death in these same tissue types after the intradermal administration of a F. oxysporum CE that was grown under the same conditions.⁴¹ These results demonstrated the potential permeation capacity through the skin and the harmful effects of this extract, even after a single application.

Table 2

Terminal dUTP nick end labeling immunostaining score of the epidermis and dermis 3, 6, 12, and 24 h after the topical application of the Fusarium oxysporum CE or CZ culture medium on the skin of Wistar rats.

Fig. 8

Rat skin photomicrography 3 h after the topical application of the CZ medium (a, c) or F. oxysporum CE (b, d). (b) Terminal dUTP nick end labeling was observed in epidermal keratinocytes (black arrow) and dermal fibroblasts (red arrow). Note the lower frequency of labeling in the control group (a). (c and d) for proliferating cell nuclear antigen immunostaining, note the higher frequency of cells stained (brown stain) on the epidermis of the animals from the group treated with the CE compared to the control group (d).

The capacity of apoptosis induction on basal cells within the epidermis after a topical application of the Fusarium T2 toxin was shown in a model of Wistar rats WBN/ILA-Ht.⁴² Albarenque and Doi⁴³ found a 40% decrease in the viability of keratinocytes in primary culture after 3 h, with evolution up to 12 h after the administration of T2 toxin. These authors used an in vitro model. Morphological alterations (e.g., chromatin condensation and margination and nuclear fragmentation, all of which are characteristics of cells undergoing apoptosis) were also recorded by optical and electron microscopy. In particular, the potential of programmed cell death induction has been attributed to trichothecenes,^42–46 which are molecules that have molecular weights of 250 to 500 Da.⁴⁷ In the present study, using a dialysis membrane with a molecular weight cut-off of 12 to 16 kDa may have removed the trichothecenes from the CE; thus, the induction of apoptosis could be attributed to other extracted components with toxic potential. Another hypothesis includes the formation of complexes between smaller molecules that can reach molecular weights $>16\mathrm{kDa}$. These complexes would not be eliminated in the dialysis process and could maintain their toxic potential, inducing the apoptosis observed by TUNEL.

It is important to observe that the period with higher intensity of TUNEL, 3 h after CE application, coincides with the most pronounced structural changes observed in the FTIR-PAS data, revealed by the modifications in the optical absorption bands associated with the ${─\mathrm{CH}}_3$ group, at ${1370\mathrm{cm}}^{-1}$. A similar result was detected in the controls, suggesting that there were structural changes that could be related to the process of programmed cell death in the skin treated with the fungal extract. Marder et al.⁴⁸ reported that the optical absorption band at ${1374\mathrm{cm}}^{-1}$ was related to the presence of mycotoxins produced by B. sorokiniana.

Table 3 and Figs. 8(c) and 8(d) show the results obtained after the immunohistochemistry reaction to detect anti-PCNA-positive cells. The results showed that there is a thickening of the epidermis within 6, 12, and 24 h after the application of the CE. The strong labeling of PCNA, the expression of which increased in the G1 and S phases of the cell cycle,^49–52 indicates an increase in the proliferative activity of this tissue after 3 h and moderate labeling after 6, 12, and 24 h. However, as shown above, strong TUNEL was observed at 3 h, and a moderate level of labeling was observed after 6, 12, and 24 h. We believe that the epidermis thickening would reflect the changes in the kinetic activity of the keratinocytes, i.e., stimulating the proliferative activity because of an increased cell death after the topical application of the extract. As shown before, the skin is one of the entrance of Fusarium.¹⁶ It is able to break the epithelial barrier inducing cutaneous lesions known as fusariosis.¹⁸ Then, it is reasonable to suppose that the topical application of the CE may be able to temporarily modify the epithelial permeability resulting in cutaneous dehydration because of the higher concentration gradient of the CE in the dermis, with consequent changes in its thickness, as occurred 3 and 6 h after the extract application. In the same period, a number of keratinocytes showed hydropic changes, a reversible state resulting from cellular injury.⁵³

Table 3

Proliferating cell nuclear antigen immunostaining scores for the epidermis and dermis 3, 6, 12, and 24 h after the topical application of the F. oxysporum CE or CZ culture medium on the skin of Wistar rats.

The skin response after topical application differed from that observed after the intradermal inoculation of the F. oxysporum CE. Although the number of inflammatory cells in the dermis increased, there was no formation of typical inflammatory infiltrate. The most frequent inflammatory cells were macrophages, followed by a small number of lymphocytes and neutrophils. These characteristics suggest that the epidermis has exercised its barrier function, preventing the extract from fully permeating.

The dermal fibroblasts were largely responsible for the increased cellularity after the application of the Fusarium extract. Fibroblasts are the cells responsible for the production of extracellular matrix components. In vitro studies showed that these cells are susceptible to the action of mycotoxins.^53,54 We observed that the dermal fibroblasts showed strong labeling for cell death after 3 h and moderate labeling in other periods. In contrast, the fibroblasts showed moderate positivity for proliferation at 3 and 6 h, a peak at 12 h, and decay at 24 h. Thus, these results indicate the pursuit of tissue component replacement after the imbalance caused by an external agent. The morphometric analysis of the dermis thickness and the area covered by collagen support this hypothesis because after 12 h, there was significant dermal thickening, which was most likely due to increased collagen synthesis [Fig. 6(b)].

4.

Conclusion

In this work, we demonstrated that the F. oxysporum CE could permeate the epidermis and dermis, reaching the subcutaneous tissue, inducing apoptosis of the cells in these tissues, and causing physicochemical changes in the organic molecules located in the dermis. Considering that the animals were immunocompetent, the proliferative and morphological changes observed in this study were derived from a process of tissue recovery after the irritating stimulation ended. We believe that the procedure used in this work may be useful for future studies in immunocompromised knockout animals to investigate the dynamics of the physicochemical changes that may occur in the infected tissues.