Differential induction of CD94 and NKG2 in CD4 helper T cells. A consequence of influenza virus infection and interferon-gamma?

Abstract

Influenza A virus causes worldwide epidemics and pandemics and the investigation of memory T helper (Th) cells that help maintain serological memory following infection is important for vaccine design. In this study we investigated CD94 and NKG2 gene expression in memory CD4 T-cell clones established from the spleens of C57BL/10 (H-2(b)) and BALB/c (H-2(d)) mice infected with influenza A virus (H3N2). CD94 and NKG2A/C/E proteins form heterodimeric membrane receptors that are involved in virus recognition. CD94 and NKG2 expression have been well characterized in natural killer (NK) and cytotoxic T cells. Despite CD94 being potentially an important marker for Th1 cells involved in virus infection, however, there has been little investigation of its expression or function in the CD4 T-cell lineage and no studies have looked at in-vivo-generated Th cells or memory cells. We show in this study that in-vivo-generated CD4 Th1 cells, but not Th2 cells, exhibited full-length CD94 and NKG2A gene expression following activation with viral peptide. For NKG2A, a novel 'short' (possibly redundant) truncated isoform was detectable in a Th2 cell clone. Another member of the NK receptor family, NKG2D, but not NKG2C or E, was also differentially expressed in Th1 cells. We show here that CD94 and NKG2A may exist as multiple isoforms with the potential to distinguish helper T-cell subsets.

Figure 1

Relative position of genes in NK cluster on mouse chromosome 6. **CD94, NKG2D, NKG2C/E and NKG2A/B/2A2, are located in the region at 130·25–130·35 Mb. Data from http://www.ensembl.org.

Figure 2

CD94 and NKG2A expression in Bpp19 and Bpp9 from total RNA. (a) Primers were designed to cover the full-length 1–732 bp translated product. Full-length NKG2A is expressed in Bpp19 but not Bpp9. (b) Primers were used that covered bp 1–656 and bp 293–656, respectively, to distinguished the isoform NKG2A from isoforms NKG2A2 and NKG2B, which would not be amplified. Positive product showed that NKG2A (bp 1–656), not NKG2A2 or NKG2B, is expressed in both Bpp19 and Bpp9. In Bpp9 Th2 cells, the NKG2A product is a non-full-length product (but covers at least the first 656 bp), designated as the ‘short’ form, whereas the Bpp19 Th1 product is part of the full-length product. (c) Primers were used to determine if Bpp19 and Bpp9 expressed NKG2D or NKG2C and NKG2E isoforms. Bpp19 expresses CD94 (as shown in a) and NKG2D, but not NKG2C or NKG2E; Bpp9 expresses NKG2A ‘short’ form (As) only (as shown in b), i.e. not NKG2C, NKG2D or NKG2E.

Figure 3

Bpp19 CD94 expression off the distal promoter. (a) The mouse CD94 gene contains six exons and codes for a transcript 848 bp in length including untranslated regions, of which 543 bp translates a 181-aa type II protein. Exons 1a and most of exon 1b encode the 5′ untranslated region, and exon 2 encodes the transmembrane domain. Exon 3 encodes the stalk region and exons 4–6 encode the carbohydrate-recognition domain (CRD). (b) Primers were used to see if a CD94 Bpp19 Th1 transcript could be expressed off the distal promoter exon 1a. The gel shows the distal primer product (636 bp) compared to the full-length product (start) (543 bp). (c) The product off the distal promoter was aligned (http://www.ensembl.org) and this showed that this isoform included a new exon, designated here as 4′. (d) Sequencing of the distal promoter isoform showed that this included a 59-bp insertion that introduces a stop codon. Arrows indicate the start of each exon. Exon 1b is coloured blue. For Bpp19 sequences, splice sites are indicated in orange (Bpp19 distal promoter) and in green (Bpp19 difference product). Start codon ATG at the end of exon 1 is in bold. Bpp19dif, Bpp19 difference product; Bpp19tot, Bpp19 start sequence; Bpp19dis, Bpp19 product distal promoter. (e) The Bpp19 difference and distal promoter products were converted to their proposed amino acid sequences and aligned (http://www.ebi.ac.uk). Translation of the distal promoter product would result in an amino acid with only part of the carbohydrate recognition domain.

Figure 4

CD94 and NKG2A expression in T cells. (a) PCR with full-length primers was carried out to investigate the expression of CD94 and NKG2A in other C57BL/10 (H-2^b) and BALB/c (H-2^d) Th1 clones to influenza HA. Positive product was found for all clones, but not for concanavalin A (ConA)-stimulated spleen cells; (b) Real-time quantitative PCR showed that CD94 and NKG2A expression were up-regulated in the activated Th1 memory clone Bpp19, but not in Bpp9. (c) Real-time quantitative PCR showed that CD94 and NKG2A are up-regulated in in-vitro-isolated Th1 cells. +, stimulated; –, unstimulated.