Diminished humoral responses against and reduced gene expression levels of human endogenous retrovirus-K (HERV-K) in psoriasis

Fourteen subjects with chronic, plaque psoriasis with affected body surface area > 10% and not on systemic medications were recruited from the UCSF Dermatology Department. All subjects provided written, informed consent for study participation under the approval of the local Institutional Review Boards. Five-millimeter punch biopsies were taken from the edge of a psoriatic plaque as well as from non-lesional skin located greater than 2 cm from any affected area. Twenty seven normal skin samples were obtained from healthy control surgical discard specimens. Skin samples were stored in RNALater (Ambion) at -80°C. Samples were mechanically homogenized using a Bio-Gen Pro 200 homogenizer and total RNA was extracted using the RNeasy mini kit (Qiagen) and a proteinase K digestion step was included in the manufacturer’s protocol. RNA was treated with DNAse at two steps, first during the extraction and secondly before converting RNA to cDNA. The quantity and quality of the RNA was assessed using a Nanodrop 8000 and in some cases by using an Agilent 2100 Bioanalyzer. For the ELISA, healthy donor sera were obtained from the Blood Center of the Pacific of San Francisco (n = 16). Psoriasis sera were obtained from subjects with chronic, plaque psoriasis recruited from the UCSF Dermatology Department.

HERV-K (HML-2) envelope transmemembrane protein (recTM), surface unit (recSU) and Gag capsid (recCA) recombinant proteins were obtained as previously described [24],[29]. Furthermore, a set of 164 overlapping “15-mer” HERV-K (HML-2) Gag peptides (JPT Peptide Technologies, Berlin, Germany) were used to comprehensively map the antibody response. 5 positive peptides were identified as reacting with healthy donor sera: 16-KRIGKELKQ AGRKGN (Matrix), 58-GYPGMPPAPQGRAPY (p15), 81-GVKQYGPNSPYMRTL (Capsid), 117-SIADEKARKVIVELM (Capsid) and 137-KCYNCGQIGHLKKNC (Nucleocapsid, NC).

ELISA was adapted from Michaud et al. [24]. Briefly, 96 microtiter wells plate (Nunc-Immuno Plate MaxiSorp Surface) were coated for 1 hour at 37°C with peptides at 10 μg/ml in phosphate buffer saline (PBS) or overnight at 4°C with recombinant protein (GeneArt) at 5 μg/ml in PBS. Plates were then washed 3 times with 200 μL of PBS/0.05%-Tween 20 and blocked with 100 μL of blocking buffer [PBS/2.5%-Bovine Serum Albumin (BSA)] at room temperature (RT). The samples were diluted in blocking buffer and incubated 2 h at RT in duplicates. Plates were then washed 3 times with 200 μL of PBS/0.05%-Tween 20. An anti-human IgG or anti-human IgM horseradish peroxidase (HRP)-conjugated secondary antibody was diluted (1:1000 for IgG and 1:2000 for IgM) in blocking buffer and incubated at RT for 1 hour. Plates were then washed 6 times with 200 μL of PBS/0.05%-Tween 20 and incubated for 10 minutes with 100 μL of TMB (3,3’,5,5’-tetramethylbenzidine, Invitrogen). Addition of 50 μL H2SO₄ 2 M stopped the reaction. The plates were read at 450 nm and 690 nm for the background on a plate reader. Background from 450 nm uncoated wells and PBS-BSA as negative controls was subtracted from the mean absorbance of the coated wells.

Expression of HERV-K (Gag, Pol, Env and Rec) and ERV-9 were studied using previous published and validated primer sequences [30],[31] (Additional file 1). DNAse treatment, cDNA conversion and PCR for all samples were done by Genomic Analysis Core Facility, Helen Diller Family Comprehensive Cancer Center, UCSF. Real time PCR was performed using the TaqMan chemistry on the ABI Prism 7900HT instrument (Life Technologies). PCR was conducted in triplicate with 20 μL reaction volumes of 1X Taqman buffer (1X Applied Biosystems PCR buffer, 20% glycerol, 2.5% gelatin, 60nM Rox as a passive reference), 5.5 mM MgCl2, 0.5 mM each primer, 0.2 μM each deoxynucleotide triphosphate (dNTP), 200 nM probe, and 0.025 unit/μL AmpliTaq Gold (Applied Biosystems) with 5 ng cDNA. A large master mix of the above-mentioned components (minus the primers, probe, and cDNA) was made for each experiment and aliquoted into individual tubes, one for each cDNA sample. cDNA was then added to the aliquoted master mix. The master mix with cDNA was aliquoted into a 384-well plate. The primers and probes were mixed together and added to the master mix and cDNA in the 384-well plate. PCR was conducted using the following cycle parameters: 1 cycle of 95° for 10 minutes and 40 cycles of 95° for 15 seconds, 60° for 1 minute. Analysis was carried out using the SDS software (version 2.3) supplied with the ABI 7900HT to determine the Cq values of each reaction. Cq values were determined for three test and three reference reactions in each sample, averaged, and subtracted to obtain the ΔCq [ΔCq = Cq (target) - Cq (reference)]. PCR efficiencies were measured for all custom assays and were greater than or equal to 90%.

RPLP0 (Taqman assay: Hs04189669_g1, Life Technologies) was used as the reference gene. The relative mRNA expression was measured as 2^(−ΔCq)*100 according to Schimttgen and Livak [32]. All the genes were compared independently.

Humoral responses assayed by ELISA were compared between groups using the two-tailed Mann-Whitney tests. The relative mRNA was compared among the different groups using Anova Kruskal-Wallis and Dunn’s multiple comparison tests. All tests were conducted using GraphPad Prism, version 6.00 (GraphPad Software, San Diego, CA), with the statistical significance of the findings set at a p value of less than 0.05.

We first investigated the humoral immune response against HERV-K in the sera of psoriasis patients and healthy controls. For these experiments, we examined the antibody responses against the two envelope sub-units encoded by the Env gene: the transmembrane protein (Env-recTM) and surface unit (Env-recSU), and the capsid encoded by Gag (Gag-recCA).

In healthy donors, we observed a basal level of IgM reacting with both Env-recSU and Gag-recCA while the response against Env-recTM was low. Interestingly, the IgM response was significantly decreased in psoriatic sera against all three recombinant proteins (Figure 1A). In healthy donor sera, the IgG response against HERV-K Env (SU and TM) was very low or absent, while a modest response against Gag-recCA was seen. This latter is significantly decreased in psoriasis patients in comparison to controls while the responses against Env-recSU and Env-recTM were not significantly modified in psoriasis patients (Figure 1B). Furthermore, we also examined IgG responses against five Gag peptides that we have previously identified as reacting with healthy donor sera among a set of 164 overlapping “15-mer” HERV-K (HML-2) Gag peptides (see Methods). We identified a significant increase of IgG in psoriatic sera against peptide 137, belonging to the HERV-K (HML-2) nucleocapsid, while the four other peptides did not display any significant difference in antibody levels (Figure 1C).

Together, these results demonstrated a significant reduction in IgM and IgG anti-HERV-K (HML-2) antibodies for most of the epitopes tested in psoriasis patients, suggesting the possibility that these proteins might be transcribed at a lower level in patients with psoriasis.

To investigate whether HERV antibody titers might be influenced by HERV expression levels, we performed RT-qPCR analysis on RNA obtained from lesional and non-lesional skin from a different set of psoriasis patients. We initially screened our samples for expression of 13 HERV genes representing several HERV families (HERV-K Gag, HERV-K Pol, HERV-K Env, ERV-9, HERV-K Rec, HERV-E4.1, HERV-R, HERV-H, HERV-L, HERV-W, HML-2, HML-4, HML-5) using published degenerate primer/probe sequences [30],[31]. For our screen, we pooled cDNA from 11 lesional and 4 non-lesional samples and performed efficiency testing using five-fold serial dilutions of the cDNA pool. Of the 13 genes screened, only 5 (HERV-K Gag, HERV-K Pol, HERV-K Env, HERV-K Rec, ERV-9) were efficient. The other 8 genes were either not efficient or did not meet the minimal threshold for expression in our cDNA pool and thus were not further evaluated.

We evaluated expression of the remaining five genes (HERV-K Gag, HERV-K Pol, HERV-K Env, HERV-K Rec, ERV-9) by qPCR in skin biopsies from 14 psoriasis patients (non-lesional and lesional) and 27 healthy controls. We used RPLP0 as an endogenous control, as it has been validated as a reference gene in prior studies of psoriasis [33]-[36]. Furthermore, we confirmed that RPLP0 is not differentially expressed between psoriasis and controls by performing RNA sequencing on an independent set of psoriasis patients (18 lesional and 16 healthy skin controls), with a differential expression q-value of 0.123 (not significant).

We found that psoriasis patients and controls both showed detectable expression of HERV genes but that lesional skin from psoriasis patients showed a significant decrease in expression of each gene as compared to skin from healthy controls (Figure 2A to E). There was also a significant decrease in expression between psoriasis non-lesional skin and control for the HERV-K Rec and ERV-9 genes. The HERV-K Env, Gag and Pol genes also showed a trend towards decreased expression in non-lesional skin as compared to control skin, with p values of 0.08, 0.05, and 0.4, respectively.

A limitation of this study is the moderate sample size examined. Some degree of variability was observed in mRNA expression across different genes. Independent replication of our findings in other cohorts is warranted.