Hi Lauren,

You could probably infer the fraction of a given lncRNA population that ends in a polyA tail through qPCR and/or Northern blotting. Let's call this fraction the "polyadenylation index", which is gene-specific. 

You would have to design two pairs of primers (qPCR) or two probes (NB). One primer would end in a stretch of Thymidine to selectively amplify  cDNA dervied from polyA+ lncRNAs, whereas the other pair would target the whole population of a specific lncRNA. This data can be used to derive a polyadenylation index value. 

I would recommend including controls known to have a low pA index (eg. histone mRNAs) and a high pA index (e.g. GAPDH mRNA). 

You can also definitely adopt the approach you mention : column separation, followed by some gene-specific detection tool, such as Northern blotting. See the Duronio paper attached for an example and methodological details (figure 3)

Best of luck,

Fabio

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1383579/

Perhaps the easiest route is to use polydT-beads that select polyA RNA. You will get two fractions: bound, and unbound. Run a q-RT-PCR on both fractions and look for relative abundance/enrichment. 

Poly A+ selection on oligo-dT is a classic method for making mRNAs.  Also PCR using a dT primer and an upstream primer will amplify the 3' end.  Look for the poly A addition signals AAUAAA in the sequence about 20-30 bases upstream of the poly A addition site.