Transforming growth factor ,Bisoforms inbreast cancer

Sir-Thestudy byMacCallum etal.(1994) onexpression of TGF-Pisoforms inbreast cancer further highlights the methodological problems associated withdetermining the roles ofthese multifunctional peptides inbothtumourprogression andresponse totherapeutic interventions. Itis important todistinguish between these twoaspects ofTGF-P function, andthis mayserve topartially reconcile someof thedisparate opinions ontheputative roles ofTGF-Pisoformsinbreast cancer. Theissue oftheorigin ofTGF-,B remains particularly controversial. Aswithprevious studies analysing total RNA extracted fromtumour samples, thedataofMacCallum etal. (1994) fail topermit anydistinction between epithelial and stromal sources ofTGF-PmRNA.Theoverexpression of TGF-Pinanautocrine capacity bybreast epithelial cells is difficult to reconcile withitsinhibitory effects upon epithelium, andtheassociation ofepithelial TGF-P1 expression withdisease progression (Gorsch etal., 1992) maybea consequence ofdefective secretion ofTGF-,B bycarcinoma cells. Nonetheless, suchsources ofTGF-Pmay serve primarily topromote stromal expansion (including angiogenesis) andhencetumourgrowth, especially inmore advanced stages ofcarcinogenesis. However, intheearlier andpremalignant stages, TGF-Pmayactpredominantly as anepithelial growth inhibitor viabothparacrine influences fromstromal cells anddirect autocrine inhibition from epithelial sources ofTGF-,B. Inaddition, TGF-Pmnay inhibit endothelial proliferation inthese earlier lesions (Schultz & Grant, 1991). Thoughimmunohistochemical studies haverevealedminimalintracellular staining of stromal cells (McCuneetal., 1992), this mayreflect thenuances ofsecretiondynamics. Moreover, whatever theroleofstromal sources ofTGF-Pintumour progression, theymayconstitute atarget forstimulation oflocal levels ofinhibitory growth factors. Inastudy fromthis laboratory (Butta etal., 1992), weobserved minimal staining ofstromal cells inpretreatment samples, butintracellular staining offibroblasts wasclearly evident following tamoxifen treatment, inaddition tomarked up-regulation ofextracellular TGF-P(between andaround stromal cells). We havealsorecently foundthatprimary cultures ofbreast tumourfibroblasts area richsource of TGF-Pl andthatlevels ofsynthesis canbemodulated by tamoxifen (ourunpublished data). Suchtherapeutic induction ofTGF-P(beitfromstromal orepithelial sources) must bedistinguished fromgrowth factor status relating toneoplastic progression perse. Itistherefore necessary toinvestigate tumours notonlyat various stages ofpresentation, butalsobefore andafter treatment interventions. Levelsof TGF-13shouldbe accurately localised andquantified. Theauthors concede that their study ispurely qualitative, andthatlevels ofTGF-,B mRNA maynotaccurately reflect levels ofprotein product owingtopost-transcriptional regulation (Knabbe etal., 1987; Colletta etal., 1990,1991; Kim etal., 1992). However, immunohistochemical studies reveal nodifferences inthepatternofexpression ofTGF-Pisoforms between benign and malignant breast tissue (Schultz & Grant, 1991), suggesting that differential quantitative expression isfunctionally important. Eventinyamounts ofTGF-PmRNA couldyield a positive signal using theRNAseprotection assaymethod described inthis paper. Furthermore, somefunctional redundancymayexist within theTGF-Pfamily, withoneisoform being pre-eminent underparticular circumstances. Weawait withinterest theresults ofimmunohistochemical andinsitu hybridisation studies.