Interactome of the oncogenic ΔNp73α isoform in human papillomavirus 38 E6/E7-transformed keratinocytes

TL;DR

This study identifies proteins that interact with ΔNp73α, an oncogenic form of p73, in HPV38-transformed skin cells linked to non-melanoma skin cancer.

Contribution

The paper presents the first interactome of ΔNp73α in HPV38 E6/E7-transformed keratinocytes.

Findings

ΔNp73α interacts with multiple proteins in both organelle and cytoplasmic compartments.

The interactome includes proteins involved in cell survival and proliferation pathways.

Findings highlight potential therapeutic targets in HPV-related skin cancer.

Abstract

ΔNp73α is a major oncogenic isoform of tumor suppressor p73. Here, we report ΔNp73α protein-binding partners in the organelle and cytoplasmic compartments of a cellular model consisting of human keratinocytes transformed by the E6 and E7 oncoproteins of the β-HPV38 virus associated with non-melanoma skin cancer.

Figures

Volcano plot analysis of ΔNp73α partners in the nucleus/organelle (left) and cytoplasm (right) fractions of 38HK cells. The black continuous line indicates the limit of significance (FDR 5%, S0 = 1). Proteins situated on the right-hand side of this line (positive Log2 value) are significant interactants. These proteins are color coded according to the localization retrieved from the Gene Ontology Resource (https://geneontology.org). Red: nucleus; green: cytoplasm; blue: mitochondria; gray: localization not reported; black: ΔNp73α bait.

Tables

TABLE 1. Top interactants of ΔNp73α (Log2 FC >2.1)a,b

Log2(FC)	-Log(P-value)	Accession	Gene name	Log2(FC)	-Log(P-value)	Accession	Gene name
Nuclear/organelle fraction
6.93	3.77	P09525	ANXA4	2.42	1.66	Q14257	RCN2
6.4	5.92	O15350	ΔNp73α	2.4	3.87	P28715	ERCC5
6.09	4.02	Q9H3D4	TP63	2.38	2.13	Q96IZ0	PAWR
5.26	2.51	P33121	ACSL1	2.38	2.35	Q13470	TNK1
4.97	4.06	P33121	MRPS22	2.33	1.36	P67775	PPP2CA
4.66	3.05	O95232	LUC7L3	2.33	1.68	Q8N9B5	JMY
4.58	4.48	P82673	MRPS35	2.32	1.31	Q9UI12	ATP6V1H
4.37	2.38	P50995	ANXA11	2.32	0.79	Q96RN5	MED15
4.03	1.98	Q92665	MRPS31	2.31	3.74	Q13573	SNW1
3.99	2.19	Q9Y2R5	MRPS17	2.3	3.34	O94915	FRYL
3.87	2.46	P27540	ARNT	2.3	1.4	Q9UHI6	DDX20
3.82	2.47	Q12965	MYO1E	2.28	5.03	Q6UB35	MTHFD1L
3.82	2.95	Q9Y3D9	MRPS23	2.28	1.78	Q14241	ELOA
3.8	3.74	O95425	SVIL	2.28	2.61	Q9UPN3	MACF1
3.78	2.51	O60828	PQBP1	2.27	1.36	O00443	PIK3C2A
3.77	1.22	Q96A65	EXOC4	2.26	4.77	P19013	KRT4
3.74	4.06	Q92552	MRPS27	2.26	1.2	Q58WW2	DCAF6
3.7	2.14	Q8IY21	DDX60	2.26	2.54	O15460	P4HA2
3.57	3.61	P82933	MRPS9	2.26	1.46	P82932	MRPS6
3.57	3.3	Q460N5	PARP14	2.25	1.47	Q9GZL7	WDR12
3.56	3.81	O43166	SIPA1L1	2.24	1.47	Q99959	PKP2
3.5	1.42	Q709C8	VPS13C	2.24	3.03	Q15427	SF3B4
3.41	1.67	Q9NP97	DYNLRB1	2.23	1.65	O43837	IDH3B
3.35	2.6	P62841	RPS15	2.22	1.7	Q9BU76	MMTAG2
3.27	2.32	O60573	EIF4E2	2.21	1.88	P56182	RRP1
3.22	3.45	Q9NYL9	TMOD3	2.21	3.37	Q96SU4	OSBPL9
3.22	1.61	Q9H788	SH2D4A	2.19	2.76	Q9UID3	VPS51
3.21	2.64	P82675	MRPS5	2.19	3.26	O75962	TRIO
3.14	2.31	Q07002	CDK18	2.19	3.77	Q5VSL9	STRIP1
3.14	2.55	O75676	RPS6KA4	2.19	2.22	Q8IXW5	RPAP2
3.12	3.12	P08758	ANXA5	2.18	4.65	Q86Y56	DNAAF5
3.12	3.72	Q9C037	TRIM4	2.18	3.88	Q96ED9	HOOK2
3.06	2.35	Q9BQS8	FYCO1	2.17	1.4	Q13144	EIF2B5
3.05	2.89	Q9BSJ2	TUBGCP2	2.16	1.67	Q99683	MAP3K5
3.04	1.53	A6NKD9	CCDC85C	2.12	1.78	Q15345	LRRC41
2.95	1.15	Q9Y3D3	MRPS16	2.11	3.81	Q14012	CAMK1
2.93	3.73	O00743	PPP6C	2.1	1.92	Q9NQG5	RPRD1B
2.89	5.13	P58107	EPPK1	2.1	2.29	Q15746	MYLK
2.88	1.46	P49841	GSK3B	2.1	1.83	P43686	PSMC4
2.87	2.04	Q6YN16	HSDL2	Cytoplasmic fraction
2.87	1.36	Q7RTP6	MICAL3	9.3	6.53	O15350	ΔNp73α
2.85	1.73	P55084	HADHB	8.8	3.45	Q9H3D4	TP63
2.84	1.56	Q15628	TRADD	4.31	3.12	O95816	BAG2
2.83	2.27	P35244	RPA3	4.22	3.05	P14735	IDE
2.8	1.12	O95639	CPSF4	4.18	2.47	Q9H4H8	FAM83D
2.8	1.75	P15735	PHKG2	4	3.29	O95232	LUC7L3
2.79	2.86	P82930	MRPS34	3.66	2.6	Q53RT3	ASPRV1
2.78	4.02	Q9BY89	KIAA1671	3.51	2.49	P33121	ACSL1
2.77	3.34	Q8N5S9	CAMKK1	3.49	3.67	Q9Y285	FARSA
2.75	2.12	Q13395	TARBP1	3.45	1.94	O14579	COPE
2.74	2.43	P51398	DAP3	3.19	3.75	Q7Z4S6	KIF21A
2.73	2.05	O43615	TIMM44	3.11	2.48	P01036	CST4
2.73	1.77	Q9UHR6	ZNHIT2	3.09	4	P53618	COPB1
2.69	2.26	Q15404	RSU1	2.82	3.01	Q8IVH4	MMAA
2.69	3.09	P30038	ALDH4A1	2.77	1.68	Q13200	PSMD2
2.69	2.04	Q9UG63	ABCF2	2.65	2.18	P12273	PIP
2.69	4.46	Q08380	LGALS3BP	2.65	1.48	P63098	PPP3R1
2.69	4.45	O15296	ALOX15B	2.58	2.16	Q6IQ49	SDE2
2.67	1.69	Q9BVI4	NOC4L	2.57	1.72	P36952	SERPINB5
2.65	0.84	Q9Y6Q9	NCOA3	2.52	4.98	Q9BPW8	NIPSNAP1
2.64	1.19	P49411	TUFM	2.47	2.37	Q9BYE4	SPRR2G
2.64	1.19	Q9BQG0	MYBBP1A	2.47	1.86	O94822	LTN1
2.62	2.09	Q9UNQ2	DIMT1	2.46	0.98	Q9C075	KRT23
2.61	2.23	O95816	BAG2	2.42	2.4	Q9BRT2	UQCC2
2.6	2.5	P55209	NAP1L1	2.36	1.96	P11310	ACADM
2.58	2	Q99700	ATXN2	2.32	1.86	Q99961	SH3GL1
2.54	3.44	Q96II8	LRCH3	2.27	2.18	P07858	CTSB
2.53	1.51	Q13613	MTMR1	2.24	2.86	P46199	MTIF2
2.52	0.81	P46019	PHKA2	2.23	4.89	P11142	HSPA8
2.51	1.87	P78347	GTF2I	2.21	2.38	P49590	HARS2
2.5	2.08	Q5T5U3	ARHGAP21	2.21	3.14	O75323	NIPSNAP2
2.47	1.12	Q9NRA8	EIF4ENIF1	2.18	2.41	P58107	EPPK1
2.46	3.36	Q8WUM0	NUP133	2.16	3.91	Q15418	RPS6KA1
2.46	2.57	Q5GLZ8	HERC4
2.46	3.09	Q86X10	RALGAPB
2.43	2.68	O43148	RNMT

Full text

ANNOUNCEMENT

The tumor suppressor p53 family comprises three transcription factors (TFs), p53, p63, and p73, which, despite sharing high sequence identity, display specific functions and regulation. All three TFs are enriched in the nucleus, where they act on transcription and DNA repair, whereas their levels are regulated in the cytoplasm by ubiquitin-dependent or ubiquitin-independent degradation. In addition, mitochondrial localization has been reported for p53 and p73 (1). The genes coding for the three proteins (TP53, TP63, and TP73) can also be expressed as multiple isoforms (2). The highly carcinogenic ΔNp73α isoform investigated here is a dominant-negative inhibitor of full-length p53 and p73 (TAp53 and TAp73α) (3). It is generated by transcription from the internal P2 promoter of TP73 and, consequently, lacks the N-terminal transactivation domain (2, 4, 5). Expression of viral oncoproteins can also result in ΔNp73α accumulation (6, 7), including the E6 and E7 oncoproteins from cutaneous beta (β) human papillomavirus (HPV) types (8–10) associated with non-melanoma skin cancer (11).

We have performed affinity purification-mass spectrometry (AP-MS) to characterize the ΔNp73α interactome using human keratinocytes immortalized by the β-HPV38 E6 and E7 oncoproteins as a model of cellular transformation (named 38HK hereafter) (8). To this end, we generated two stable cell lines, expressing either ΔNp73α fused to the N-terminus of the tandem affinity purification (TAP) tag (ΔNp73α-TAP) or TAP alone (negative control) (12). To discriminate between cytoplasmic versus nuclear or other organelle interactants, 38HK ΔNp73α-TAP and TAP 38HK cells were fractionated as described in our previous report (12). Briefly, 38HK cells were lysed mechanically in hypotonic buffer and centrifuged. The supernatant, corresponding to the cytoplasmic extract, was recovered. The pellet, containing nuclei and other organelles, was resuspended in buffer containing NaCl (250 mM) and glycerol (20%), incubated on ice, and centrifuged. The supernatant from this second centrifugation corresponds to the nuclear/organelle extract. Then, extracts were processed by tandem affinity purification (13) following the protocol described in Reference (12). Protein complexes were eluted from calmodulin beads and analyzed by SDS-PAGE and Western blotting or silver staining. Protein levels were quantified by the Bradford assay.

Eluates from calmodulin beads were precipitated with trichloroacetic acid, and the pellet was washed with acetone and dissolved with 2 M urea. Samples were digested with 100 ng of trypsin at 37°C for 10 h. The resulting peptide mixtures were directly analyzed by MS with a nanoLC U3000 in-line coupled with an Orbitrap Elite mass spectrometer using a nano-electrospray source (Thermo Scientific). MS measurements were performed on one biological sample per condition. Technical variability was estimated from three MS acquisitions for each sample (technical replicates). Peptides were separated using a C18 column (75 µm × 25 cm) with a 35 min linear gradient from 5% to 50% buffer B (A: 0.1% trifluoroacetic acid in H_2_O/B: 80% ACN, 0.08% trifluoroacetic acid in H_2_O). The mass spectrometer was operated in data-dependent mode with survey scans from m/z 300–1650 acquired in the Orbitrap at a resolution of 120,000 at m/z 400. Proteins were identified using Proteome Discoverer 2.5 software (Thermo Scientific), with false discovery rate (FDR) < 1% and at least one unique peptide. Label-free quantification was based on eXtracted Ion Chromatography (XIC) using the Minora node from Proteome Discoverer. Precursor and fragment mass tolerance were set at 7 ppm and 0.5 Da, respectively. Oxidation (M) was set as variable modification and carbamidomethylation (C) as fixed modification. The human fasta database with 20,419 sequences was obtained from Uniprot (https://www.uniprot.org/proteomes/UP000005640). Statistical analyses were performed using Perseus 1.6.1.5 according to the processing workflow described by the authors (14). The repeated t-test used for volcano plot analyses was based on standard parameters of Perseus (FDR 5%, S0 = 1).

The lists of the top interactants of ΔNp73α (Log2FC > 2.1) in the nucleus/organelle and cytoplasmic fractions are reported in Table 1. The statistics from Perseus is presented in a volcano plot with each protein colored according to Gene Ontology Cellular Component (Fig. 1). The nuclear/organelle and cytoplasmic fractions are strongly enriched with protein partners with reported nuclear and cytoplasmic localization, respectively, thereby validating our biochemical approach. While more abundant in the nuclear/cytoplasmic fractions, mitochondrial binders are also present in the cytoplasmic fraction.

Volcano plot analysis of ΔNp73α partners in the nucleus/organelle (left) and cytoplasm (right) fractions of 38HK cells. The black continuous line indicates the limit of significance (FDR 5%, S0 = 1). Proteins situated on the right-hand side of this line (positive Log2 value) are significant interactants. These proteins are color coded according to the localization retrieved from the Gene Ontology Resource (https://geneontology.org). Red: nucleus; green: cytoplasm; blue: mitochondria; gray: localization not reported; black: ΔNp73α bait.

In our previous study (12), we have characterized the interaction of ΔNp73α with the E2F4/p130 transcriptional repressor complex, which is detected in the nuclear/organelle fraction of these analyses. Results show that E2F4 binds to ΔNp73α but not to the full-length TAp73α protein, suggesting that the interactomes of the oncogenic isoform and of the tumor suppressor protein may be different.