Localization of mRNAs by in-situ hybridization to the residual body at stages IX-X of the cycle of the rat seminiferous epithelium: fact or artefact?

Abstract

Several recent articles have reported localization of specific mRNAs in the rat testis to stage IX and X seminiferous tubules using in-situ hybridization. In all cases the expression was located basally in the tubules and appeared as discrete round clusters of grains close to the lamina propria. The localization was interpreted as being in Sertoli cells or leptotene spermatocytes. In this study we demonstrate that this pattern is most probably due to artefactual binding of probes to the residual body (RB). In the present study testicular tissue, perfusion-fixed with Bouin's and embedded in paraffin, was used, as this resulted in excellent morphological preservation such that RBs within tubules at stages VIII-X were clearly distinguishable. RNA content of the RBs was demonstrated at stages VIII-X using methyl green pyronin staining, and could be eliminated by pretreatment with RNAse or trichloroacetic acid. Localization of mRNAs for 11 seminiferous tubule proteins was assessed using 35S-labelled and digoxigenin-labelled riboprobes (activin receptor-II, alpha-inhibin, transferrin, androgen-binding protein (ABP), cyclic protein-2 (CP-2), CREM, sulphated glycoproteins 1 and 2 (SGP-1 and SGP-2), transition protein 2 (TP-2) and cystatin-C), and digoxigenin-labelled oligonucleotide probes (transition protein-1 (TP-1), TP-2 and protamine-1). All of these probes showed localization to the correct cell type(s) within the seminiferous epithelium. In addition, six antisense riboprobes (activin receptor-II, CREM, SGP-2, CP-2, cystatin C and alpha-inhibin) showed hybridization to basally located residual bodies in tubules at stages IX-X on one or more occasions, whereas residual bodies around the edge of the lumen (stage VIII) or in transit through the seminiferous epithelium showed no hybridization; sense probes showed no localization to residual bodies. A common feature of the probes which localized to the basal RBs was that they had been prepared using cDNA cloned into Bluescript SK- vector such that the antisense strand was generated from the T7 polymerase promotor. A cRNA prepared using T7 polymerase and Bluescript vector alone and a GC-rich 27mer oligonucleotide corresponding to the region of the multiple cloning site of Bluescript adjacent to the T7 site both localized uniquely to basal RB. It is concluded that the hybridization seen within RBs is probably a subtle artefact unique to RBs undergoing dissolution following fusion with Sertoli cell lysosomes, and may reflect nonspecific hybridization to GC-rich fragments of RNA.