How does tetracycline transactivator localize to the nucleus? Where is the nuclear localization signal?

Question

Proteins typically use a nuclear localization signal (NLS) to localize to the nucleus. Tetracycline transactivator (TTA) needs to work in the nucleus, but I did not find an NLS in the structure.

Tetracycline transactivator is a fusion of tetracycline repressor of E. coli (P04483) and VP16 of HSV (although by blast I only found P04483).

The Tet-Off system makes use of the tetracycline transactivator (tTA) protein, which is created by fusing one protein, TetR (tetracycline repressor), found in Escherichia coli bacteria, with the activation domain of another protein, VP16, found in the herpes simplex virus.2. Source

There can be multiple NLS's, but the ones I checked I did not find here.

Elements

1. TetR is a protein from E. coli, which has no nucleus, and the above sequence is a perfect match to P04483, so NLS is not expected here.
2. VP16 is from HSV, and it is reported to use HCF-1 to locate to the nucleus.

VP16 itself does not possess a nuclear localization signal; it is transported to the nucleus by HCF-1 and then binds to Oct-1, which can also recognize the same target sequence. - Since only a domain of VP16 is (f)used in TTA, I am not sure if this is present.

3. Vaysse 2013 explicitly added an SV40 NLS to TetR, although for a completely different reason (translocate the plasmid DNA in to the nucleus during transfection).

Sequence of Tetracycline transactivator (advanced)

tccagactggacaagagcaaagtcatcaactctgccttggagctcctgaatgaagttggcattgagggcttgaccaccaggaagctggcccagaagctgggtgtggagcagcctaccctgtactggcatgtgaagaacaagagggctctgcttgatgccctggccattgagatgttggacaggcaccacacccacttctgccctctggaaggggagtcctggcaggacttcctgaggaacaatgccaagagcttcagatgtgccttgctctcccaccgggatggtgccaaagttcacttgggcaccaggcctacagagaagcagtatgagaccctggagaaccagctggcattcctgtgccaacaaggcttctccctggaaaatgccttgtatgccctctctgctgtgggccacttcaccttgggctgtgtgctggaggaccaggagcaccaagttgccaaggaggagagggagacccccaccactgactccatgccaccactgctgcggcaagctattgagttgtttgaccaccaaggggctgagcctgcattcctttttggcctggaactgatcatctgtggcctggaaaagcagctgaaatgtgagtctggctctcacgtgcccaaaaagagaaagcacgtgcctgctgatgccttggatgattttgacctggacatgctgcctgctgatgccctggatgactttgatttggacatgctccctgctgatgcacttgatgattttgacctggatatgctg

Answer 1

Proteins with a molecular weight less than ~30 kDa(*) can diffuse freely through the nuclear pore (see for example here (1) for a review on nuclear transport).

Assuming that the sequence you provide for tTA is correct (EDIT: it is not, see edit below), the protein would have a molecular weight of 28.64 kDa, i.e., below the size limit. In other words, it seems like an NLS is not essential for tTA entry into the nucleus, which is evident from this paper (2) where they use tetracycline to regulate expression from a tTA-dependent promoter in HeLa cells.

Having said that, in this paper (3) (unfortunately paywalled) they create a tTA-nls derivate of tTA by fusing an NLS (KRPRP) to its N-terminus. No data on nuclear localization is shown in the paper itself, but they refer to a thesis (4) where nuclear enrichment of tTA-nls has apparently been shown.

So, although tTA can enter the nucleus without an NLS, including one will cause it to actively localize into the nucleus, possibly improving its transcription factor activity.

(*) This is a number I've heard before, and apart from finding it in the review I haven't bothered looking up the original reference. There are more accurate ways (5) of defining the free diffusion barrier, but I'll stick to ~30 kDa for the purpose of this answer.

EDIT: I did a bit more research, and it looks like the DNA sequence you provide is incorrect. The first part does encode for TetR (2-207), but the second part is not something I could identify, but it is certainly not VP16 (see EDIT 2 below).

In the original paper (2), two versions of tTA are created. In both cases they use an N-terminal region of TetR (1-207):

>TetR (P04483) 1-207 MSRLDKSKVINSALELLNEVGIEGLTTRKLAQKLGVEQPTLYWHVKNKRALLDALAIEMLDRHHTHFCPLEGESWQDFLRNNAKSFRCALLSHRDGAKVHLGTRPTEKQYETLENQLAFLCQQGFSLENALYALSAVGHFTLGCVLEDQEHQVAKEERETPTTDSMPPLLRQAIELFDHQGAEPAFLFGLELIICGLEKQLKCESGS

They then use two different fragments of VP16 and fuse them C-terminally directly to the above TetR fragment:

1. Long VP16 fragment (363-490)

   >VP16 (P06492) 363-490 AYSRARTKNNYGSTIEGLLDLPDDDAPEEAGLAAPRLSFLPAGHTRRLSTAPPTDVSLGDELHLDGEDVAMAHADALDDFDLDMLGDGDSPGPGFTPHDSAPYGALDMADFEFEQMFTDALGIDEYGG

2. Short VP16 fragment (406-490)

   >VP16 (P06492) 406-490 HTRRLSTAPPTDVSLGDELHLDGEDVAMAHADALDDFDLDMLGDGDSPGPGFTPHDSAPYGALDMADFEFEQMFTDALGIDEYGGF

   The trailing phenylalanine (F) in the second sequence is not part of the original VP16 sequence, but gets added as a result of their method of cloning.

As far as I can see, they only use the longer version of tTA in the paper, and show that it works in HeLa cells. This is despite the fact that this protein actually has a molecular weight of 36.92 kDa. This is above the ~30 kDa size limit I mentioned earlier, but since this limit is only approximate and tTA clearly works, I think it is still safe to say that tTA freely diffuses through the nuclear pore (albeit not as efficiently as a smaller protein might).

EDIT 2: after the Op provided the reference for the DNA sequence (GenBank: DQ414432.1), I was able to identify the ORF from this source paper (6) as "improved tTA" (itTA), and the C-terminal fragment as three F-domains (PADALDDFDLDML).

References

1. Kim YH, Han M-E, and Oh S-O (2017) PMCID: PMC5509903.
2. Gossen M and Bujard H (1992) PMID: 1319065.
3. Gossen et al (1995) PMID: 7792603
4. Kistner A (1993) thesis, University of Heidelberg, Germany.
5. Matsuda A and Mofrad MRK (2021) PMID: 34339633.
6. Krestel HE et al (2004) PMID: 15548671.