The official website of the HHMI Science Education Alliance-Phage Hunters Advancing Genomics and Evolutionary Science program.

Welcome to the forums at Please feel free to ask any questions related to the SEA-PHAGES program. Any logged-in user may post new topics and reply to existing topics. If you'd like to see a new forum created, please contact us using our form or email us at

Refining the call for Cas4 family exonuclease vs. RecE-like exonuclease

| posted 07 May, 2021 15:56
I'd like some community input on how everyone is thinking about the call of Cas4 family exonuclease vs. RecE-like exonuclease. According to the approved functions list, we call RecB-like exonuclease/helicase if we can confidently identify both a helicase and an exonuclease domain present in the protein. Cas4 family exonuclease is called if we see alignments to the crystal structure 3H4R_A and to the PD-(D/E)XK nuclease superfamily (PF12705.7, among others) according to the functions list. It's a little more unclear on calling the RecE-like exonuclease. Presumably we should be looking for alignments to RecE crystal structures. I've been using this as a guide for my own annotations.

Using Che9c_60 as a model for RecE-like exonuclease (from approved list), HHPred reveals matches to 3H4R_A and PF12705.7. Che9c_61 is called as a RecT-like ssDNA binding protein. There are no alignments to anything labeled as RecE.

My question is: should we be calling RecE-like exonuclease for a gene that has the exonuclease only domains and is immediately upstream of a RecT-like ssDNA binding protein (regardless of hits to 3H4E_A and PD-(D/E)XK nuclease superfamilies), and reserve Cas4 family exonuclease for genes that are not immediately upstream or very close to a RecT-like ssDNA binding protein? My immediate thoughts are to use the presence of RecT as a guide to call RecE-like exonuclease, but I want to hear what you all think.
| posted 07 May, 2021 19:23
Hi Kristen and all,
This is one of those genes that needs a dig deep. Here are some of what i have been thinking.
I am reluctant to call a RecE without a RecT.
I don't know how to differentiate between a RecE, a generic exonuclease or the Cas4 family exonuclease. I am looking forward to what we can figure out about these.
Good question, sorry to not be of much help.
| posted 20 May, 2022 17:06
I have done some analysis of RecE while annotating G1 phage ShaboiShabazz gp42. First, the PDB hit to 3H4R_A are to an E. coli exodeoxyribonuclease VIII, this is the putative RecE gene and has no connection to Cas4. Second, the publication regarding this RecE gene has identified several “conserved” domains and the genes within the PHAM that includes ShaboiShabazz gp42 contain 4 of these domains with near 100% conservation of not only the key amino acids, but also the secondary structures (B sheets and alpha helices). Third, RecE (exodeoxyribonuclease VIII) function in concert with the RecT. RecE binds to free double-stranded DNA(dsDNA) ends and processively digests the 50-endedstrand to form 50-mononucleotides and a 30-over-hang that is a substrate for single strand annealing promoted by RecT (2009: Structure 17, 690–702). ShaboiShabazz gp43 hits to PF03837.17: RecT family. Its key to note that RecT functions in BOTH RecA-dependent and RecA-independent DNA recombination pathways, so having a hit to this Pfam is not specific to Cas4 and I’m not sure how these connections were even made. **SEE THE POSTED ANALYSIS OF gp42 showing the clear hits to RecE.

For Pfam 12684.10: This entry represent a PD-(D/E)XK endonuclease-like domain superfamily. PD-(D/E)XK nucleases constitute a large and highly diverse superfamily of enzymes that display little sequence similarity. However, they share a common core fold and a few critical active site residues and ARE NOT specific to Cas4 either. Makes sense for RecE to hit to this Pfam based on overall function.

Thus, hits to 3H4R_A SHOULD CALL RecE and having a downstream RecT also confirm that calling both genes is valid. Thus: ShaboiShabazz gp42 is RecE-like exonuclease and gp43 is RecT-like ssDNA binding protein.
RS Pollenz
Edited 20 May, 2022 17:07
| posted 20 May, 2022 19:48
| posted 17 Aug, 2022 17:29
Rick, Debbie, Kristen,

Not all members of this pham have the RecT-like gene immediately downstream. Should those also be called RecE-like anyway? There are (at least) two patterns:

1. For cluster P2 phage Tortellini gp55, the next gene is not RecT, but has weak hits to a similar ssDNA binding protein (see attached phamerator map and HHPred results for Tortellini gp56) The rest of the cluster P phages do have the RecE-RecT combination described for ShaboiShabazz.

2. In some cluster AY phages (Persistance gp62, Seahorse gp72), there is a ssDNA binding protein 3-4 genes downstream (Persistance gp65, Seahorse gp76). a) Is this close enough to be identified as a RecE-RecT-like? b) HHPred hits for the ssDNA binding protein (Persistance_65) do not specifically mention RecT but rather M.smeg SSBb and related proteins.

It is challenging to distinguish between RecE-like exonuclease, RecB exonuclease domain, and Cas4 family exonuclease based just on alignments, since they are so similar. I was investigating cluster P1 phage Langerak, which has a gene in this pham (gp46). HHPred alignments to all three types of hits are in the attachment. They all include the same motifs that Rick identified for RecE in ShaboiShabazz gp42 and many of the specific residues. One could argue that there is better alignment to RecE (there are larger gaps in alignments to RecB-like and Cas4) – is this be sufficient to call RecE in the absence of RecT? Does it come down to specific residues? Or is a the more general exonuclease call the better option in the absence of RecT?

Login to post a reply.