Q7VZI5 (BP0922) — PTHR43709 comparison (2026-10-21)

Script: compare_prpf_family.py (run with uv run --script compare_prpf_family.py);
raw output in results.json. Sequences and PANTHER/NCBIfam cross-references were
fetched live from UniProt REST. Identity is from a global BLOSUM62 alignment,
counted over the shorter sequence.

Percent identity of Q7VZI5 to PTHR43709 members

Accession	Protein	PANTHER SF	NCBIfam	% id to Q7VZI5
Q7VW94	B. pertussis BP2370, prp-operon PrpF	SF2	TIGR02334	36.1
Q8EJW4	S. oneidensis PrpF (characterised)	SF2	TIGR02334	36.8
Q937N7	C. necator PrpF	SF2	TIGR02334	36.6
Q0QLE6	E. barkeri Mii, 3-methylitaconate isomerase (characterised)	SF2	-	43.7
A0A0U2X0E4	U. maydis Adi1, aconitate delta-isomerase	SF2	-	35.5
O07931	B. subtilis YraM	SF2	-	39.2
P0AAV8	E. coli YbhH	SF3	NF033377	39.4
Q0KJL4	Novosphingobium LigU, oxalomesaconate delta-isomerase	SF3	NF033377	44.8
Q88JY0	P. putida GalD, 4-oxalomesaconate tautomerase	SF3	NF033377	44.3

The three PrpF proteins (TIGR02334 matches) are 78-80% identical to one another,
including the B. pertussis prp-operon paralog BP2370. Q7VZI5 is about 36% identical
to all of them and does not match TIGR02334. Its identity to every characterised
member of the family, in either subfamily, falls between 35% and 45%, with no clear
nearest neighbour. Mii is at 43.7%, and LigU and GalD (SF3) are at 44-45%.

Active-site residues of S. oneidensis PrpF mapped onto each sequence

These are PrpF Lys73, Cys107, Lys281 and His317, which contact trans-aconitate in the
PrpF crystal structure (PMID:17567742).

Accession	K73	C107	K281	H317
Q7VZI5	K65	C99	R266	H301
Q7VW94	K70	C104	K278	H314
Q0QLE6 (Mii)	K62	C96	M263	H300
SF3 members	K	C	K	H

The proposed catalytic Lys/Cys pair is conserved across the family, Q7VZI5 included.
This is consistent with a DapF-fold isomerase. One carboxylate-binding lysine is an
arginine in Q7VZI5 and a methionine in Mii. The family keeps the catalytic machinery
but varies its substrate-binding residues.

Genomic context (UniProt REST, proteome UP000002676)

B. pertussis carries a canonical prp cluster: BP2368 prpC, BP2369 aconitate
hydratase (AcnD-type), BP2370 PrpF (TIGR02334), BP2371 prpD.
BP0922 sits in a different cluster. Its neighbours are BP0918 (MmgE/PrpD-family
protein), BP0919 (succinate-semialdehyde dehydrogenase), BP0920 (exported
FAD-binding protein), BP0921 ("citrate utilization protein B"), BP0923
("phosphosulfolactate synthase"-like), BP0924 (transcriptional regulator) and BP0925
(fumarylacetoacetate-family hydrolase).

Interpretation

Q7VZI5 is not the PrpF ortholog of B. pertussis. BP2370 is. Q7VZI5 is a divergent
paralog inside PTHR43709:SF2, a subfamily that also contains enzymes with different
substrates (PrpF, Mii, Adi1).
Nothing in these data assigns a substrate. A methylitaconate delta-isomerase
(Mii-like) activity cannot be transferred, because 43.7% identity sits within the
same range as the unrelated SF3 tautomerases. Propionate catabolism cannot be
transferred either, because the methylcitrate-cycle isomerase role belongs to BP2370.
What is supported is a family-level activity: an isomerase of the DapF fold with
the conserved catalytic Lys/Cys pair. The substrate is unknown.
Limitations: this is pairwise analysis only. No tree was built and no PAINT node
placement was inspected for Q7VZI5.

Hydropathy of Q7VZI5

Kyte-Doolittle, 19-residue window: the maximum over the whole sequence is 1.52 and
the maximum over the first 40 residues is 0.55. No window passes the usual
transmembrane threshold of about 1.6, and the N-terminus is not hydrophobic. There is
no sign of a signal peptide or a transmembrane segment. That fits a soluble protein,
in the same way the soluble, crystallised PrpF protein of S. oneidensis is one. This
is a crude screen, not SignalP or Phobius.