EVALUATION OF FIXATIVE SOLUTIONS FOR ULTRASTRUCTURAL ANALYSIS OF BROWN SPIDER LOXOSCELES INTERMEDIA ( ARANEAE : SICARIIDAE ) TISSUES

In view of the wide variety in the composition of fixative solutions used for spiders, five different fixative formulae were tested for fixing male brown spider (Loxosceles intermedia) gonad tissues. The brown spider represents a public health problem in the city of Curitiba. The morphological study of its gonads may aid in the understanding of its reproductive strategies and possibly in the development of a growth control program. The fixatives tested contained glutaraldehyde alone or combined with paraformaldehyde, and the buffers cacodylate or phosphate, with or without the addition of sucrose or sodium chloride as osmolytes. The fixative containing 2.5% glutaraldehyde and 2% paraformaldehyde, in 100 mM phosphate buffer with 200 mM sucrose, or in 200 mM sodium cacodylate preserved well mitochondria, Golgi apparatus, and membranes in general. These formulas were nearly isosmotic (439 mOsm/kg H2O e 455 mOsm/kg H2O respectively) to the brown spider hemolymph (478 mOsm/kg H2O). Concerning the fixative agents, the combination of glutaraldehyde and paraformaldehyde is advised to give optimal fixation of Loxosceles intermedia cells. For other species of spiders, hemolymph osmolality should be considered, but the fixative formulae cited above will also probably give good results.


INTRoDuCTIoN
The genus Loxosceles includes the most poisonous spiders in Brazil.In the urban environment of the city of Curitiba, capital of Paraná State, the brown spider L. intermedia Mello-Leitão -1934 (Araneae, Sicariidae) (Platnick -2004) is the most common species, specially in domestic habitats where -according to recent reports -it poses an increasing danger.The poison of L. intermedia acts proteolytically (Silveira et al., 2002), causing a local necrotic skin lesion, frequently with systemic effects that can be lethal (Gonçalves de Andrade et al., 2000).The study of L. intermedia reproduction and its morphological features by electron microscopy can promote better understanding this spider's biology and reproductive strategies.Furthermore, it should be useful in developing a reproduction control program.Considerable variation in composition of buffer and fixative solutions for studying spider tissues is found in ultrastructural studies in the literature.It is well known that a detailed analysis of any tissue or organ's ultrastructure requires optimal preservation of the material.Therefore, fixatives, buffer compositions and osmolality, and type of osmolytes should all be carefully considered before being selected.
This work reports treatment of male brownspider gonads: the tissue chosen, and the five different fixative solutions tested, all of which were of different osmolalities and buffer compositions.

Material analyzed
Adult brown-spider (Loxosceles intermedia) males were donated by inhabitants Curitiba, Paraná State,Brazil (25° 25' 40'' S and 49° 16' 23'' W).The spiders were kept in small plastic vials in the laboratory, and twice a month received Tenebrium sp.larvae and water in a small, saturated cotton ball).A total of 15 adult male L. intermedia specimens were used, along with 3 spider testes fixed with each of the 5 different fixative solutions tested.Spider hemolymph was extracted from the cephalothorax at the point where the legs are inserted.Two hemolymph pools, each from 5-6 spiders, were obtained.

Fixative solutions
The compositions of the 5 fixative formulas tested appear in Table 1.In every case, the pH was 7.4.Fixative solutions (FS) were prepared with glutaraldehyde alone (FS 4) or glutaraldehyde plus paraformaldehyde.Glutaraldehyde was tested in two different concentrations (2% in FS 1, and 2.5% in FS 2 -5).Phosphate (FS 1, 2, and 3) and sodium cacodylate (FS 4 and 5) were tested as buffers, and osmolytes tested were NaCl (FS 1) and sucrose (2).Phosphate buffer was first tested as an nontoxic alternative to cacodylate.Measured osmolality of the fixative solutions was 423-523 mOsm/kg H 2 O above buffer osmolality, except for fixative 4, in which the difference was 211 mOsm/kg H 2 O.

Osmolality measurements
Osmolality of the two hemolymph pools and of all fixative solutions and buffers was measured using a vapor pressure osmometer (VAPRO 5520, Wescor, USA).Each sample was measured in triplicate.

Dissection, fixation, and embedding method
Testes dissection, which took about 15 min/ spider was done with the spider immersed in the fixative solution chosen.Fixation, including dissection, required 3 h and was done at 4 °C, followed by several washings in the buffer being tested.Following post-fixation with 1% osmium tetroxide, prepared with the same buffer used in the primary fixative, samples were washed in distilled water, dehydrated in a graded series of ethanol and embedded in Spurr's resin, according to standard protocols.

Sectioning, staining and analysis
Ultrathin sections (70 ηm) were obtained using a Leica Ultracut ultramicrotome (UCT), and were contrasted with 5% uranyl acetate for 30 min, and for 10 min with Reynold's lead citrate.Sections were examined in a transmission electron microscope (TEM), jeol JEM-1200 EX II; electron micrographs were obtained with Gatan Digital Micrograph software for image acquisition.The analyses were done by comparing tissue preservation -particularly of the mitochondria, the Golgi apparatus, and the cellular membranes -based on the electron micrographs.

ReSuLTS
The mean osmolality of the two pools of brown spider hemolymph was 478 mOsm/kg H 2 O.As described in Table 1, buffer solution 1 (100 mM phosphate buffer plus 50 mM NaCl) was hyposmotic to brown spider hemolymph.Buffer solution 2 (100 mM phosphate buffer plus 200 mM sucrose) was slightly hyposmotic, and the others (solution 3: 200 mM phosphate buffer, and solutions 4 and 5: 200 mM sodium cacodylate) were nearly isosmotic to hemolymph osmolality.
Table 2 summarizes the results obtained with the five fixative solutions (FS).In the electron micrographs of the testes fixed with FS 1, the cells appeared swollen, despite the addition of mM NaCl to the phosphate buffer.Mitochondria presently frequently swollen cristae (Fig. 1); the images of the Golgi apparatus were poorly defined.
The tissue treated with FS 2 presented clear indications of swelling and poor preservation, despite the addition of 200 mM sucrose.The cells, however, displayed intact, much more clearly defined cytoplasmic membranes, and their mitochondria presented a less dense matrix and were better preserved than the membranes themselves.But cristae swelling was still evident (Figs. 2 and 5).
As for FS 3, the use of 200 mM phosphate buffer resulted in completely dried out tissue, which fragmented when sectioned (Table 2).
With respect to FS 4, which contained 2.5% glutaraldehyde (without paraformaldehyde) and 200 mM sodium cacodylate instead of phosphate, the result was slightly better tissue preservation, as compared to the results of all previous fixative solutions.Only occasional swelling was observed, and cells and organelles were generally well preserved (Figs. 3 and 6).However, this solution made the cytoplasm appear flocculated.The best result was obtained using FS 5, which contained  2.5% glutaraldehyde with the addition of 2% paraformaldehyde in 200 mM sodium cacodylate buffer (Figs. 4 and 7).The tissue appeared uniform, and presented little swelling or tissue disruption; cell membranes were intact, mitochondria and the Golgi apparatus very well preserved, with a clear matrix, intact membranes, and only occasional swelling (Table 2).

DISCuSSIoN
This study recounts a search for a suitable buffer and fixative solution for ultrastructural study of Loxosceles intermedia using male gonad tissue.In the literature on spider ultrastructure, a wide variety exists in buffer and fixative solution composition.Solutions buffered with either phosphate (Suzuki & Kondo, 1994;Suzuki, 1995;Michalik et al., 2004), or sodium cacodylate (Uhl, 2000) in different concentrations, with or without salt addition, have been used for various species of spiders.
In the present study, five different fixative formulae were prepared and tested, based on the formulas described in the above references, and modified in order to adapt them to brown spider hemolymph osmolality, which is 478 mOsm/kg H 2 O and within the 400-600 mOsm/kg H 2 O range for spiders cited by Foelix (1996).
With respect to osmolality, buffer 1 (100 mM phosphate buffer plus 50 mM NaCl) was not adequate, probably because it was hyposmotic (302 mOsm/kg H 2 O) to the brown spider hemolymph (Tables 1 and 2, Fig. 1).The other buffer solutions were nearly isosmotic to the hemolymph.However, not only is osmolality significant, but also the buffer and fixative solution composition (Hayat, 1970).An ionic osmolyte was used in FS 1 (NaCl), and a non-ionic osmolyte in FS 2 (sucrose), which rendered the osmolality value (439 mOsm/kg H 2 O) close to that of the hemolymph, and yielded better results in comparison to those of FS 1 (Tables 1  and 2, Figs. 2 and 5).The successful association of phosphate buffer and sucrose (Suzuki & Kondo, 1994;Michalik et al., 2004) is confirmed in the present work, in which this combination stabilized membranes of brown spider cells (Figs. 2 and 5).
Better results with nonionic osmolytes, such as sucrose or dextran, than NaCl have already been reported (Hayat, 1970).As for FS 2, although it produced well-defined membranes, it was unsatisfactory in preserving mithocondria, in which some tissue swelling was observed.
Cacodylate -used as the buffer (200 mM) in FS 4 and FS 5 (Table 1), with no osmolyte added -was nearly isosmotic to the brown spider hemolymph, and produced good results (Table 2;Figs. 3,4,6,and 7).At this same concentration (200 mM), phosphate buffer in FS 3 (Table 1) resulted in unsectionable tissue in these 3 preparations.Thus, cacodylate stands out as a better buffer than phosphate for spider testes (Hayat, 1970).Phosphate at a concentration of 100 mM, in association with 200 mM sucrose, should be considered in studying brown spider tissues when use of toxic agents such as cacodylate is ruled out.
Use of glutaraldehyde with paraformaldehyde in FS 5 (Tables 1 and 2; Figs. 4 and 7) resulted in better preservation of the testes tissue, when compared to the results of glutaraldehyde alone (FS 4: Tables 1 and 2; Figs. 3 and 6), with which cytoplasmic organelles were visible, but the cytosol seemed coagulated.With FS 5, the cytosol appeared more uniform, and all membranes were perfectly preserved (Tables 1  and 2; Figs. 4 and 7).Glauert (1975) reported that the glutaraldehyde and paraformaldehyde combination results in better tissue preservation than either alone, because paraformaldehyde penetrates tissues much more rapidly and stabilizes the structures, while glutaraldehyde produces more permanent fixation.Our results with FS 4 and FS 5 on brown spider tissue confirm Glauert`s observation.
The present results clearly evidenced that for optimal tissue preservation, attention should be paid to the extracellular fluid osmolality of animals being studied, and to the aldehydes added, buffer (cacodylate or phosphate), osmolytes added (sucrose or NaCl), as well as to the total osmolality of the solution.The fixative formulas successfully used in the present work for ultrastructural studies of brown spider testes could also be tried for the tissues of other spider species, specially if their hemolymph is osmotically similar to that of Loxosceles intermedia.