Field production and functional evaluation of chloroplast-derived interferon-alpha2b

Abstract

Type I interferons (IFNs) inhibit viral replication and cell growth and enhance the immune response, and therefore have many clinical applications. IFN-alpha2b ranks third in world market use for a biopharmaceutical, behind only insulin and erythropoietin. The average annual cost of IFN-alpha2b for the treatment of hepatitis C infection is $26,000, and is therefore unavailable to the majority of patients in developing countries. Therefore, we expressed IFN-alpha2b in tobacco chloroplasts, and transgenic lines were grown in the field after obtaining United States Department of Agriculture Animal and Plant Health Inspection Service (USDA-APHIS) approval. Stable, site-specific integration of transgenes into chloroplast genomes and homoplasmy through several generations were confirmed. IFN-alpha2b levels reached up to 20% of total soluble protein, or 3 mg per gram of leaf (fresh weight). Transgenic IFN-alpha2b had similar in vitro biological activity to commercially produced PEG-Introntrade mark when tested for its ability to protect cells against cytopathic viral replication in the vesicular stomatitis virus cytopathic effect (VSV CPE) assay and to inhibit early-stage human immunodeficiency virus (HIV) infection. The antitumour and immunomodulating properties of IFN-alpha2b were also seen in vivo. Chloroplast-derived IFN-alpha2b increased the expression of major histocompatibility complex class I (MHC I) on splenocytes and the total number of natural killer (NK) cells. Finally, IFN-alpha2b purified from chloroplast transgenic lines (cpIFN-alpha2b) protected mice from a highly metastatic tumour line. This demonstration of high levels of expression of IFN-alpha2b, transgene containment and biological activity akin to that of commercial preparations of IFN-alpha2b facilitated the first field production of a plant-derived human blood protein, a critical step towards human clinical trials and commercialization.

Figure 1

Confirmation of chloroplast integration and determination of homoplasmy/heteroplasmy in the T₀ generation of both varieties. (a) The 0.81-kb DNA probe containing chloroplast flanking sequences. (b) DNA fragments of 7.9 kb indicate untransformed chloroplasts and DNA fragments of 9.9 kb are observed when the chloroplast genome contains the integrated transgenes. Lanes 1, 5, untransformed wild-type; lanes 2–4, transgenic LAMD-609 lines; lanes 6–9, transgenic Petit Havana lines. (c) Western blot of LAMD-609 transgenic plants expressing chloroplast-derived interferon-α2b (cpIFN-α2b). Low-nicotine tissue extract separated by 15% sodium dodecylsulphate-polyacrylamide gel electrophoresis (SDS-PAGE), with cpIFN-α2b detected by mouse monoclonal antibody against human IFN-α. Lane 1, 80 ng PEG-Intron™ standard; lane 2, protein marker; lane 3, wild-type (untransformed) LAMD-609; lanes 4–7, transgenic LAMD-609 lines expressing monomers and multimers of cpIFN-α2b. (d) Western blot of Petit Havana transgenic plants expressing cpIFN-α2b. Petit Havana leaf extract separated by 15% SDS-PAGE, with cpIFN-α2b detected by mouse monoclonal antibody against human IFN-α. Lane 1, 38 ng of Intron®A; lane 2, protein marker; lane 3, 190 ng of Intron®A; lane 4, wild-type (untransformed) Petit Havana; lanes 5–7, transgenic Petit Havana lines expressing monomers and multimers of cpIFN-α2b; lanes 8, 9, Escherichia coli-transformed with IFN-α2b.

Figure 2

Quantitation of interferon-α2b purified from chloroplast transgenic lines (cpIFN-α2b). Crude extracts of transgenic leaf material were analysed for IFN-α2b expression by indirect enzyme-linked immunosorbent assay (ELISA). Quantitation of transgenic IFN-α2b was performed on the T₁ generations of LAMD (a, b) and Petit Havana (c, d) plants. (a, c) Quantitation of cpIFN-α2b as a percentage of total soluble protein (TSP). (c, d) Quantitation of cpIFN-α2b as milligrams of transgenic protein per gram fresh leaf weight. Two different standards were used in the ELISA: recombinant IFN-α2b with a covalent conjugate of monomethoxy polyethylene glycol (PEG-Intron™) and recombinant human IFN-α2b (rHuIFN-α2b). Y, young leaves (top few in the plant); M, mature leaves (fully developed); O, old leaves (bottom senescent leaves with decreased pigments). Duration of illumination: 0, 1, 3 and 5 days of continuous illumination. Error bars represent the standard error of the mean.

Figure 3

Transgenic tobacco plants produce interferon-α2b (IFN-α2b) with biological activity in vitro. (a) Transgenic IFN-α2b inhibits vesicular stomatitis virus (VSV)-induced cytopathicity. Baby hamster kidney (BHK) cells were pretreated with the indicated samples for 24 h prior to VSV exposure. At the conclusion of the assay, the cells were washed and stained with crystal violet for microscopic examination. (b, c) Transgenic IFN-α2b inhibits human immunodeficiency virus (HIV) infection, as determined by a decrease in luciferase expression. TZM-BL cells were pretreated for 24 h with the indicated reagents, and then exposed to two different HIV isolates: the CCR5-tropic strain BaL (b) and the CXCR4-tropic strain IIIB (c). All treatments were performed in triplicate. Error bars represent the standard error of the mean. Asterisks indicate P < 0.05.

Figure 4

Immunohistochemical detection of the interferon response in mouse spleens. (a) Representative photographs of mouse spleen tissue. The tissues were fixed and 10-µm sections were stained for either the natural killer (NK) cell markers CD49b and NK1.1 or major histocompatibility complex class I (MHC I). For NK cell marker staining, the channels were read individually and also overlaid. It should be noted that, because of the natural expression profile of MHC I, all cells are presumed to be positive. Therefore, these data are designed to evaluate relative differences between the conditions. Isotype controls depict the background fluorescence of the native tissue. Photographs were taken with a 20× objective and are representative of more than 100 sections of each treatment condition. (b, c) Analysis of the mean fluorescence intensity (MFI) of splenocytes stained for MHC I-positive cells (b) and NK cells (c). All tissue samples were acquired with equivalent MFI. Error bars in (b) and (c) represent the standard error of the mean. Asterisks indicate statistical comparisons between PEG-Intron™ and phosphate-buffered saline (PBS), and interferon-α2b purified from chloroplast transgenic lines (cpIFN-α2b) and PBS, with significant P values of less than 0.05.

Figure 5

Phenotypic analysis of splenocytes. Total splenocytes isolated from the various groups of mice in this study were analysed by flow cytometry for surface marker expression. (a) Representative flow cytometry plots showing splenocyte expression of major histocompatibility complex class I (MHC I) and CD49b. Quadrant positions are based on isotype controls. (b) Comparison of the mean fluorescence intensity (MFI) of MHC I across the various treatment groups. All cells comprising the MHC I-positive population (i.e. both CD49b-positive and CD49b-negative populations) are represented. (c) Comparison of the expression of natural killer (NK) cell markers (CD49b and NK1.1) across the various treatment groups. The data in (b) and (c) represent the means of each treatment group. Error bars represent the standard error of the mean. Asterisks indicate statistical comparisons between PEG-Intron™ and phosphate-buffered saline (PBS), and interferon-α2b purified from chloroplast transgenic lines (cpIFN-α2b) and PBS, with significant P values of less than 0.05.

Figure 6

Transgenic tobacco plants produce interferon-α2b (IFN-α2b) with biological activity in vivo. (a) Photographs of lungs of mice injected with the metastatic tumour B16-F10. Mice were treated via intraperitoneal injection, as indicated, for three consecutive days. (b) Mean colour intensity of tumour metastases. Photographs of lung samples were analysed for colour intensity. All samples were photographed with equal exposure settings. A calculation of total colour was then achieved by selecting only those pixels that corresponded to the lung tissue. The mean colour intensity was calculated for all samples of a particular group. The lower the intensity, the greater amount of black (tumour) in the image. Error bars represent the standard error of the mean. Asterisks indicate statistical comparisons between PEG-Intron™ and phosphate-buffered saline (PBS), and purified cpIFN-α2b and PBS, with significant P values of less than 0.05.