Retroviral sero-reactivity in LGL leukaemia patients and family members

Abstract

T-cell large granular lymphocyte (T-LGL) leukaemia is characterized by a clonal proliferation of cytotoxic T cells and is frequently associated with rheumatoid arthritis. Sera from some LGL leukaemia patients react to a portion of the human T-cell leukaemia virus (HTLV-1/2) transmembrane envelope protein, BA21, although HTLV-1/2 infection is rare in LGL leukaemia patients. Here we show that family members, including spouses, of an LGL leukaemia patient had elevated LGL counts, BA21 reactivity and, additionally, recognition of HIV-1 gp41. Thus, both LGL leukaemia patients and clinically normal contacts sharing the same environment have evidence of exposure to a retrovirus.

Keywords: epidemiology; leukaemia; retroviruses.

Figure 1.. Retroviral antigen reactivity of LGL leukaemia patients.

A.) Seroreactivity Summary Table; left: Western blot of retroviral lysates, right: enzyme-linked immunosorbent assay (ELISA) reactivity to human immunodeficiency virus (HIV) recombinant protein. Levels of total IgG (all classes) and retrovirus-reactive IgG were determined by densitometric scan of Western blots for sera from 24 normal donors, 24 LGL leukaemia patients and control donors infected with each virus (Pos CTL): 16 HIV-1, 16 HTLV-1, 4 HIV-2, 4 HTLV-2, and from 2 Rous Sarcoma Virus (RSV) immunized rabbit sera. The asterisk (*) indicates that HIV-1 infected control sera were used because no bovine leukaemia virus (BLV)-infected control sera were available. Retrovirus lysates were prepared from gradient-purified viruses. On the right panel, serological reactivity of these groups to recombinant viral proteins from HIV-1 Env, HIV-1 Gag, and HIV-1 Pol were assessed by ELISA. All OD450 values were obtained from the mean corrected OD450 values of each donor group and were directly proportional to the amount of reactive serum IgG. Standard deviations are included in parentheses under each experimental value. The ELISA results were analysed for significance using the non-parametric paired two-tailed T test. B.) Custom peptide arrays were generated to cover HIV-1 MN Env and a consensus sequence of HIV-1 subgroup B p55 Gag. Shown are data from a representative experiment, which included HIV-infected donors (HIV+, n=3), normal donors (NORM, n=3) and LGL leukaemia patients (LGL, n=6). The thermogram shows peptide signal intensities in red (+4) vs. green (−4). PEP# represents the National Institute of Health AIDS Research and Reference Reagent Program ID number of overlapping peptide. The adjusted intensity values from LGL leukaemia and HIV-infected sera were standardized against normal donor values to calculate normalized signal intensities for each peptide. Mean normalized signal intensities are listed under the heading for each serum group. Non-reactive peptides are included to illustrate non-reactive flanking sequences. Peptide 6326 was included to show typical low LGL leukaemia serological reactivity to gp120 peptides. Red print sequences: the region of overlap between adjacent peptides. p17: Matrix, p24: Core Antigen, NC: Nucleocapsid, sig: gp160 polyprotein signal leader sequence, gp120: outer membrane glycoprotein, gp41: transmembrane glycoprotein. BA21-LIKE: Sequences aligning with HTLV-1 BA21, shown in fig 1C. C.) Alignment of HIV-1 gp41 BA21-LIKE sequence with HTLV-1 BA21. Highlighted sequence shows regions of homology.

Figure 2.. LGL cell counts and human retrovirus serology in four generations of a family of an LGL leukaemia patient.

Genealogy of the patient’s family. The LGL leukaemia patient is indicated (LGL, red circle). Squares denote males, circles females, and diagonal arrows indicate decedents. The grandmother of the patient had seven husbands; the two shown indicate the husband who is the grandfather of the patient and the current husband, who was included in this study. I: LGL counts in family members of a patient with T-LGL leukaemia. Increased numbers of T- and NK-LGLs (> 3 standard deviations from normal controls using 95% confidence interval test), as determined by flow cytometry, are denoted by darkened circles and squares. N: Cell count data was not available for this specimen. II: Seroreactivity to HTLV-1 BA21 is indicated by darkened circles and squares. Reactivity to other HTLV antigens is indicated by a dot in the centre of a circle or square. N: Seroreactivity data was not available for this specimen. III: The HIV Env seroreactivity shown here is to the VGGLVGLRIVFAVLSI portion of Env that corresponds to the amino terminus of BA21 and is indicated by darkened circles and squares. The family circled in blue is the patient’s brother, his wife and his sons, who all had increased LGL counts and were also reactive to the BA21 and the HIV BA21-like epitopes.