Figure 1.
Schematic drawing of the filamentous phage structure.
(A) The wt virion is made up of five structural proteins that coat a single stranded DNA genome of about 6.4 kb. (B) In the wt phage there is about 2,700 copies of pVIII and approximately 3–5 copies each of the four proteins pIII, pVI, pVII and pIX, which are found at each tip of the virion [4], [35]. The virion size depends on the genome size at approx. 2.3 nucleotides per pVIII, and hence the length of the particle changes as a function of genome length [36]. The theoretical MW of the mature capsid proteins were calculated from the sequence of VCSM13 (GenBank accession no.: AY598820).
Figure 2.
Schematic drawing of the pVII and pIX display phagemids.
The full-length pIII of pSEX81 (GenBank accession no.: Y14584) was exchanged with either full-length pVII (N-MEQVADFDTIYQAMIQISVVLCFALGIIAGGQR-C) or pIX (N- MSVLVYSFASFVLGWCLRSGITYFTRLMETSS-C) retrieved from the M13K07 genome giving rise to pGALD7 (A and B) and pGALD9 (C and D), respectively. Moreover, both pGALD7 and pGALD9 were made such that the recombinant fusion was targeted either to the periplasm by a pelB signal sequence (A and C), or not (B and D). The phagemids can accommodate in-frame exogenous sequences (e.g. variable gene segments from Abs creating single chain Fv (scFv)) introduced through cassette exchange on NcoI/HindIII or MluI/NotI, respectively, thereby creating a continuous open reading frame (ORF). The two cassettes are connected by a 21aa synthetic linker containing the mAb Yol1/34 tubulin epitope (bold underlined) [18]. The heterologous polypeptide is fused to the capsid through a 9aa spacer containing a trypsin protease site (bold underlined). Unique restriction sites are indicated. Abbreviations: lacPO, lac promoter; SD, Shine-Dalgarno sequence; pelB, signal sequence of bacterial pectate lyase; t, T7 transcriptional terminator.
Figure 3.
Evaluation of pIII, pVII and pIX mediated display of folded domains.
(A) The cell density (A600 nm) of the individual phagemid packaging cultures was measured (as 1/4th dilutions and the undiluted OD back-calculated) after ON growth at 30°C. (B) Phagemid titers in culture supernatants were determined by infectious titration. Transcription of the POI-capsid fusions, controlled by lacPO, was at basal level only after removal of repressor (glucose) from the growth medium (uninduced), or also induced by addition of 1 mM IPTG (IPTG induction). (C – E) Samples of cleared supernatant after phagemid rescue were used to assess antigen reactivity by phage capture ELISA. mAbs F23.2 and GB113 are conformation specific and clonotypic for the scTCR, respectively. NC is the anti-M13 detection Ab only and an irrelevant scTCR (7A10B2) was included as a negative control (♦). (F) The phagemid to helper phage ratio for the samples in B shown as cfuampR/cfukanR based on infectious titration.
Figure 4.
Agarose gel of virion ssDNA content and infectious titration analysis.
(A) Equal volumes of PEG precipitated M13K07-rescued scFv anti-phOx display samples were separated on a 1% agarose gel, the virions denaturated and the ssDNA content visualized as described in Methods. The titer in-puts were not normalized; hence the band intensities directly reflect the ssDNA type and amount. The larger sized band is the helper phage genome, whereas the smaller sized bands are the phagemids. The M13K07 helper phage was included as a control (K). The actual phagemid/helper phage genome sizes are: pGALD7ΔL, 3679 bp; pGALD7, 3739 bp; pGALD9ΔL, 3679 bp; pGALD9, 3739 bp; pSEX81, 4882 bp and M13K07, 8669 bp. (B) The phagemid to helper phage ratio for the samples in A shown as cfuampR/cfukanR based on infectious titration. (C) The primary infection titers in cfu/ml of the samples in A, which were used to determine the ratios depicted in B.
Figure 5.
POI-capsid fusion display level determined as antigen specific binding.
Serial dilutions of phagemid-rescued samples displaying either scFv anti-phOx (A), or scFv anti-NIP (B) were applied in a phage capture ELISA. Antigen-bound virions were detected by anti-M13-HRP and the data shown as a function of the number of virions applied per well. For all samples, the virion titer was determined by absorbance at A268 nm and hence does not discriminate between phagemid or helper phage containing virions.
Table 1.
Infectious output (cfuampR) after monoclonal mock selection.
Figure 6.
Antigen-specific enrichment in affinity selection depends on capsid display scaffold.
Two rounds of affinity selection were performed using two different scFv specificities displayed on either of pIII, pVII or pIX. In each library, a total of 12, the specific scFv was spiked into a large background of virions (1∶107) as described in Methods. All selections were done in parallel on both antigens (phOx-BSA and NIP-BSA) employing three different virion elution conditions following antigen binding; high pH (TEA), direct in-well infection, or proteolytic antigen-virion disruption (trypsin). Following amplification, equal volumes of virion-containing culture supernatants from selection round 1 and 2 were assessed for antigen reactivity by phage capture ELISA. Round 0 corresponds to the spiked input of 1×1010 cfuampR. To estimate maximum possible response (100% enrichment of the specific scFv), culture supernatants from the pure specific virions that were spiked into the libraries were used as reference for the corresponding selection. The results are given as fraction of the maximum, indicated by cone shape. Importantly, the virion titers of all supernatants were roughly equal (data not shown).