A chloroplast transgenic approach to hyper-express and purify Human Serum Albumin, a protein highly susceptible to proteolytic degradation

Abstract

Human Serum Albumin (HSA) accounts for 60% of the total protein in blood serum and it is the most widely used intravenous protein in a number of human therapies. HSA, however, is currently extracted only from blood because of a lack of commercially feasible recombinant expression systems. HSA is highly susceptible to proteolytic degradation in recombinant systems and is expensive to purify. Expression of HSA in transgenic chloroplasts using Shine-Dalgarno sequence (SD), which usually facilitates hyper-expression of transgenes, resulted only in 0.02% HSA in total protein (tp). Modification of HSA regulatory sequences using chloroplast untranslated regions (UTRs) resulted in hyper-expression of HSA (up to 11.1% tp), compensating for excessive proteolytic degradation. This is the highest expression of a pharmaceutical protein in transgenic plants and 500-fold greater than previous reports on HSA expression in transgenic leaves. Electron micrographs of immunogold labelled transgenic chloroplasts revealed HSA inclusion bodies, which provided a simple method for purification from other cellular proteins. HSA inclusion bodies could be readily solubilized to obtain a monomeric form using appropriate reagents. The regulatory elements used in this study should serve as a model system for enhancing expression of foreign proteins that are highly susceptible to proteolytic degradation and provide advantages in purification, when inclusion bodies are formed.

Figure 1

Integration of transgene cassettes into the chloroplast genome and study of homoplasmy. (a) Regions for homologous recombination are underlined in the native chloroplast genome. HSA is driven in all cassettes by the Prrn promoter upstream of the aadA gene for spectinomycin resistance with additional promoters and control elements as described in the text. Arrows within boxes show the direction of transcription. Numbers to the right indicate the predicted hybridizing fragments when total DNA digested with BamHI is probed with probe P1. (b) The 0.81 kb fragment (P1) flanking the cassette and 0.75 kb fragment containing HSA coding region (P2) were used as probes for the Southern blot analysis. (c, d) Southern blot analysis. 1: untransformed DNA; DNA from plants transformed with: 2,3: pLDAsdHSA; 4,5: pLDApsbAHSA. Plants for the first (T₀) and second (T₁) generation were analysed. 2,4: T₀ generation. 3,5: T₁ generation. Blots were probed with P1 (c) and P2 (d). AadA: aminoglycoside 3′-adenylyl transferase; kb: kilobases; P: promoter; Prrn: 16SrRNA promoter; SD: Shine-Dalgarno.

Figure 2

Transcription patterns of transgenic plants. A Northern blot analysis was performed with total RNA extracted from leaves of potted plants. The 3′ of the psbA gene was used as probe. 1: untransformed plant; 2: transformed with pLDAsdHSA; 3: transformed with pLDApsbAHSA after illumination or 4: in the dark. Ethidium bromide-stained rRNA was used to assess loading. Identified transcripts are indicated to the right. A scheme of transcription patterns expected for the different cassettes integrated into the chloroplast genome is shown at the bottom of the figure. Horizontal arrows above genes show anticipated transcripts. Arrows within boxes show the orientation of genes within the chloroplast genome. Read through transcripts are not shown in this figure. rRNA: ribosomal RNA.

Figure 3

Analysis of HSA accumulation in transgenic chloroplasts. (a) ELISA of HSA accumulation in leaves of potted plants at different stages of development. Samples were collected from untransformed plants or transformed with pLDAsdHSA or pLDApsbAHSA. Expression levels are indicated as a percentage of total protein. (b) Study after different hours of illumination. Samples of leaves were collected from potted plants transformed with pLDApsbAHSA after the 8-h dark period or at indicated hours in the light. (c) Coomassie stained gel to study HSA accumulation in tobacco leaves of potted plants. Total protein extracts were loaded in the gel. 1: 500 ng pure HSA; 2: molecular weight marker; 3: untransformed plant; transformed with 4: pLDAsdHSA; 5: pLDApsbAHSA after 8 h of illumination; 6: pLDApsbAHSA after 8 h of darkness. Between 40 and 50 μg of plant protein were loaded per well. The positions of HSA and RuBisCO large subunit (LSU) are marked. (d) Colorimetric immunoblot detection of tobacco protein extracts from mature leaves in potted plants. Total protein extracts were loaded in the gel. 1: 40 ng pure HSA; 2: molecular weight marker; 3,5: untransformed plant extract; 4: pLDAsdHSA plant extract; 6: pLDApsbAHSA plant extract. Between 40 and 50 μg of plant protein were loaded per well. kDa: kiloDalton; LSU: RuBisCO large subunit.

Figure 4

Study of HSA accumulation into inclusion bodies. (a–d) Electron micrographs of immunogold labelled tissues from untransformed (a) and transformed mature leaves with the chloroplast vector pLDApsbAHSA (b–d). Note presence of inclusion bodies (b–d) marked with an arrow in (d). Scale bars indicate μm. Magnifications are a × 10 000; b × 5000; c × _{6300; d} × 12 500.

Figure 5

Plant T₁ phenotypes. 1,2: untransformed plants; 3: plant transformed with pLDAsdHSA; 4: plant transformed with pLDApsbAHSA.

Figure 6

HSA extraction from inclusion bodies. (a) Silver stained SDS-PAGE gel showing 1: 500 ng pure HSA; 2: molecular weight marker; soluble fraction obtained after centrifugation of pLDApsbAHSA transformed plant extract (lane 3) or untransformed plant extract (lane 4); 5: HSA after solubilization from the pellet; 6: proteins from untransformed plant, which followed the same process as the proteins of lane 5. Amounts of protein loaded per well were 10 μg in lanes 3 and 4, 550 ng in lane 5 and 450 ng in lane 6. (b) Chemiluminiscent immunoblot detection of protein extracts. 1: 40 ng pure HSA; 2: HSA from a plant transformed with pLDApsbAHSA during the solubilization process, showing mono, di and trimeric forms; 3: proteins from an untransformed plant that followed the same process as the proteins for lane 2; 4: same HSA from lane 2 but in a more advanced stage of solubilization; 5: completely monomerized HSA after the end of the solubilization treatment (the sample of this lane corresponds with lane 5 in (a)).