Preparation of monoclonal antibody bank against ...

Preparation of monoclonal antibody bank against whole proteins from fast-growing bamboo shoots Yu-Jen Wu 1, 2, Han-Min Chen 3, Dai-Jer Wu 1, Jiann-Shing Wu 1, Rea-Min Chu 4 and Rong-Huay Juang 1* 1

Department of Biochemical Science and Technology, and Institute of Microbiology and Biochemistry, National Taiwan University, Taipei, Taiwan 106

2

Department of Beauty Science, Meiho Institute of Technology, Pingtung, Taiwan, 912

3

Department of Life Science, Catholic Fu-Jen University, Taipei, Taiwan 242

4

Department of Veterinary Medicine, National Taiwan University, Taipei, Taiwan 106

*

Address correspondence to: Rong-Huay Juang E-mail: [email protected] Phone: 886-2-3366-4448

Abstract. When the total proteins in the proteome of an organism were used as a complex antigen for immunizing a mouse, the host spleen could generate specific cell lines producing antibodies against most of the individual proteins in the proteome. Hybridoma were then produced by fusing the splenocytes with myeloma, the immortalized B-cell lines were selected and subcloned, and propagated for the production of monoclonal antibodies (First-stage mAb bank) against the antigenic proteins. The antigens in the proteome were removed by an affinity column prepared by immobilizing the First-stage mAb. Proteins flowed through the affinity column were used for the Second-stage immunization and mAb bank preparation. By using this two-stage strategy, we report here the production of an antibody bank containing 192 individual monoclonal antibodies against the proteins extracted from the fast-growing green bamboo shoots. In a relatively short period of time, this high-throughput platform produced useful antibody repertory, which might be utilized in the antibody chip preparation, or exploring the protein-protein interactions in the target proteome.

p. 1

As proteomic research brings in powerful analytical technique by using twodimensional electrophoresis (2DE) for separation, and mass spectroscopy for protein identification, there is an urgent need for specific probe to the target spots on the 2DE gel 1, 2

. Although computer software could help to some extent, it is a tedious work to compare

the protein spots on the 2DE gels, especially when the proteome are too complex to resolve the spots unambiguously. Antibody is a specific probe against its antigen, and could be produced by immunizing the antigen against an animal host of different species. Each antibody is secreted by its specific B-cell line which could be immortalized by fusing with myeloma cell; and the resulted hybridoma cell line produces single monoclonal antibody (mAb) which recognizes only its specific antigenic determinant 3. Theoretically the host will produce antibody against every invading protein which contains peptide with different amino acid sequence. Therefore, if the total proteins in a proteome of foreign tissue are used for immunization, it is possible to produce a bank of mAb against most proteins in the proteome. However, some weak immunogenic, or low-abundant proteins, might not trigger their specific mAb effectively. A strategy to produce antibodies against these poor antigens is the affinity chromatographic removal of the proteins which induce antibodies readily during the First-stage immunization. The remaining proteins which can not be recognized and adsorbed are then subjected to the Second-stage immunization to another mouse (Fig 1a). If necessary, more cycles of adsorption and immunization could be performed. To test the applicability, we prepared a mAb bank against the total proteins from the shoots of green bamboo (Bambusa oldhamii) by this multiple stage protocol. The final mAb bank is useful in investigating the expression of proteins during the rapid growth of bamboo shoots, which will be published elsewhere. The total water-soluble proteins of the bamboo shoot (60 cm height above the ground) was extracted and analyzed by 2DE (Fig. 2a). All detailed procedures are described on the Supplementary methods (online). An aliquot of the total proteins was used in immunizing two Balb/c mice on a fixed schedule; mouse sera were collected every two weeks and examined by 2DE and immunostaining. The titer started to increase after 6 weeks, and reached to the maximum at 12th week (Supplementary Fig. 1 online). The mice were sacrificed after final immunization, their spleen cells collected and fused with myeloma NS0/1 (Mouse A) or Sp2/0 (Mouse B), respectively. After fusion, each fusion mixture was distributed into three 96-well microtiter plates which were cultured p. 2

immediately under HAT (hypoxanthine-aminopterin-thymidine) selection (D0, Fig. 1b). On D14, supernatant (50 μL) from each well was sampled and analyzed by onedimensional electrophoresis (1DE, 12.5% SDS-PAGE) and immunostaining (First screening). Positive clones were picked and expanded to T-25 culture; the T-25 medium was used to reveal the binding pattern of the mAb by 2DE immunostaining (First identification). Candidate clones were selected according to the First identification, and were further subcloned by limiting dilution (one 96-well plate per candidate). Positive clones were screened again by 1DE and immunostaining (Second screening), and then transferred to T-25 for another 2DE analysis (Second identification). Duplicate clones were identified and eliminated by examining their 2DE immunostaining patterns during the two identification steps. The culture media of the final First-stage cell lines were collected individually, and purified by ion exchange chromatography (DEAE Sepharose) to obtain the IgG fractions which were then immobilized on CNBr-Sepharose for the affinity adsorption of their antigens. The proteins flowed through the affinity column were collected and used in immunizing another mouse (Mouse C) for the preparation of the Second-stage mAb bank, which followed the same protocol as the First-stage (Fig 1a). The results of the screening and selection were summarized in Table 1. After the Firststage cell fusion, about one thirds of the microtiter wells showed positive on 1DE blots (First screening). While each positive well might contain more than one cell line, the positive numbers expanded to 1.5 or 2 folds after the subcloning, and totally 400 clones were positive (Second screening). However, some clones showed identical 2DE patterns (Second identification), and only one of them was kept in the final mAb bank. Therefore, the final number of the First-stage bank shrunk to 160, in which Mouse A produced 78 individual clones, and Mouse B contributed 82 clones. For the Second stage, only 40 wells (14%) showed positive after the First screening, since the protein repertoire used as the antigen in this stage was simpler and less immunogenic. After the subcloning and Second screening, 32 individual clones were identified as distinct from the cell lines of the First stage. Totally, we obtained 192 mAb showing distinctive 2DE staining patterns (Supplementary Fig. 2 online). It is noticeable that most mAb in this bank stained more than one spot on its 2DE blot. Some unique clones (e.g. AE6B12 in Supplementary Fig. 2a online) had very complex patterns looked like stained with polyclonal antiserum. We have checked these clones further by another subcloning step, and confirmed that they were monoclonal. For an approximate estimation, about 70 mAb (36%) showed simple p. 3

staining patterns (less than three spots), and 40 mAb (21%) obtained complex patterns like AE6B12. The total number of spots recognized by the First-stage mAb bank reached to 1,140 as summing up all the 160 blots, and 220 for the Second-stage. On an average, every mAb can stain up to seven spots on the 2DE. This observation illustrates the complexity of the protein modification after its translation. Another explanation is that several independent proteins might share common antigenic determinant which could be recognized by the same mAb. To check to total coverage of protein spots recognized by mAb in the bank, the mixture of 160 mAb produced in the First stage were used for staining the total proteins on the 2DE blot. The resulting pattern (Fig. 2d) revealed only part of the total spots on the silverstained gel (Fig. 2a), since weak immunogenic or poor expressed proteins can not induce Ab efficiently. Nevertheless, some highly immunogenic proteins produced mAb with enhanced titer in this stage (indicated with horizontal arrows in Fig. 2d). The affinity adsorbent immobilized with these 160 mAb did remove a great part of the immunogens, as compared the 2DE patterns of the Second-stage antigens (Fig. 2b) with the First-stage (Fig. 2a). Accordingly, the immunostaining pattern using the Second-stage mAb bank (Fig. 2e) was very different from the First-stage bank (Fig. 2d). When we further compared the pattern stained by a mixture of all 192 mAb (Fig. 2h) with the total proteins (Fig. 2a), there were still some proteins which cannot produce antibody throughout these two stages. However, immunostaining patterns by mouse antisera collected during immunization (Fig. 2c, g & f) did stain additional spots (examples were marked with asterisks). We might lose these clones during the screening and subcloning procedure. On the other hand, some spots were stained strongly only with mAb, but not antisera (examples indicated with vertical arrows in Fig. 2d & e). In the First stage, we immunized two Balb/c mice with the same proteins for three months; the immune response was rather different as revealed from their 2DE patterns by antisera (Fig. 2c & g). Serum from Mouse B showed higher titer and more complex staining pattern than Mouse A. Their spleen cells were then fused with myeloma NS0/1 and Sp2/0, respectively. Nonetheless, these two mice produced nearly equal number of mAb in the First-stage banks (Table 1a). In this study, only the IgG-secreting cell lines were collected in the final bank, since the anti-mouse IgG second antibody was used in all screening procedures. The IgG subclass is strategically selected, because it is monomeric, and is induced in the later period of immunization, which might generate higher affinity p. 4

against the antigens through the somatic mutation and affinity maturation. Recent reviews 1, 2 have showed the urgent need of specific probes for the identification of unknown proteins in a proteome; instead of the antibody technique, phage display is generally regarded as the tool of choice for this purpose 1. The major obstacle for producing an antibody bank by the hybridoma technique is its redundant and timeconsuming work. Besides, some minor proteins are difficult to induce antibody since they are poor immunogenic or insufficiently expressed. Nevertheless, immune response in the vertebrate is still the most effective and reliable way for producing specific-binding molecules against defined targets. To our knowledge, at least one group has demonstrated the possibility of producing about 50 mAb against the mitochondrial proteins from human liver {Gao, 2006 3008 /id}. Our study stepped further using different screening strategy and produced near 200 individual mAb against proteins from bamboo shoots by a twostage procedure, which could solve the problem of the minor proteins. In total, only three mice were used in the whole experience which was completed in 18 months by two fulltime research staffs. We believe that the number of the mAb should increase if one or more stages of adsorption and immunization could be performed.

Reference List 1. Konthur, Z., Hust, M., & Dubel, S. Perspectives for systematic in vitro antibody generation. Gene 364, 19-29 (2005). 2. Hanash, S. HUPO initiatives relevant to clinical proteomics. Mol. Cell Proteomics. 3, 298-301 (2004). 3. Kohler, G. & Milstein, C. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature 256, 495-497 (1975). 4. Gao, J. et al. Proteomics-based generation and characterization of monoclonal antibodies against human liver mitochondrial proteins. Proteomics 6, 427-437 (2006).

p. 5

Table 1 Summary of the stages for mAb production (a) First-stage Spleen cell fused with Number of clones

NS0/1 (Mouse A)

Sp2/0 (Mouse B)

screened

positive

screened

positive

First screening

150

100

180

120

Second screening (after subcloning)

320

150

500

250

78

Final monoclonal

82

(b) Second-stage Spleen cell fused with Number of clones

Sp2/0 (Mouse C) screened

positive

First screening

100

40

Second screening (after subcloning)

400

120

Final monoclonal

32

p. 6

Figure Legends Figure 1 Flow charts for the stage-wise production of mAb bank. (a) Every stage consists of the immunization, fusion and screening steps for an antigen of complex proteins. The proteins which induced antibodies successfully in the First stage were removed by affinity adsorbent immobilized with all the First-stage mAb. The remaining proteins were then used as the antigen for the Second stage. We have carried out two stages only in this study. (b) The detailed procedure for screening in each stage was illustrated; an approximate time schedule was started from the day of fusion (D0). For the First screening, the one-dimensional SDS-PAGE and immunostaining was used instead of the enzyme-linked immunosorbent assay (ELISA) in traditional hybridoma screening. The binding specificity for each individual mAb was examined by 2DE immunostaining. Some clones might show identical staining patterns on their 2DE blots, only one was kept in the final bank.

Figure 2 The immune response against the complex protein antigens from the rapidgrowing bamboo shoots was monitored by 2DE and immunostaining. (a, b) The silver-staining patterns of 2DE gels for the total proteins (a, used for the First-stage antigen) and the remaining proteins after affinity adsorption (b, for the Secondstage antigen) were compared. (c-h) Mouse antisera were collected during their immunization, and used to stain the First-stage (c, g) and the Second-stage (f) antigens. On the other hand, blots stained by mAb mixtures from the First-stage bank (d) and the Second-stage bank (e) showed distinctive patterns to each other. Finally, the total 192 mAb from these two stages was mixed to reveal the whole binding pattern (h), which looked similar to the total protein pattern (a) but not identical.

p. 7

Figure 1

a

b (bamboo shoot)

Affinity column (I)

Affinity column (II)

Total Proteins

Partial proteins

Partial proteins

(antigen)

(not adsorbed)

(not adsorbed)

First-stage Immunization

Second-stage Immunization

Third-stage Immunization

Fusion & screening

Fusion & screening

Fusion & screening

mAb bank (I)

mAb bank (II)

mAb bank (III)

First stage

Second stage

(Third stage)

Starting material

p. 8

Figure 2

Total proteins (antigen for First-stage mAb bank preparation)

a

First-stage immunization

(antigen for Second-stage mAb bank preparation)

Second-stage immunization

Cell fusion and screening

Mouse A

d

c

Proteins after absorption

b

Mouse C

e

f *

* *

Mouse B

g

First-stage bank

Second-stage bank

(mixture of 160 mAb)

(mixture of 32 mAb)

Antiserum

*

h *

Antiserum Total mAb bank (mixture of 192 clones)

p. 9

Supplementary Information Methods Preparation of the total soluble proteins from green bamboo shoots Green bamboo shoots (Bambusa oldhamii) were collected from the local farm near Muzha, Taipei, Taiwan. Fresh bamboo shoot (100 g, 60 cm high from the earth) were homogenized in a Waring blender with 100 ml extraction buffer (50 mM Tris-HCl, pH 8.0, 1 mM EDTA, 20 mM DTT, 1 mM benzamidine, 2 μM leupeptin, 1 μM pepstatin), the extract was centrifuged for 30 min at 21,600 g (Beckman Avanti J25, rotor JLA 16.25) to remove the insoluble debris 1. The supernatant was collected and used directly for immunization, or subjected to the following TCA/acetone precipitation for two-dimensional electrophoresis (2DE). Protein content was determined by the dye-binding method (Bradford 1976) using the microassay system from Bio-Rad (Protein Assay Kit). Bovine serum albumin was used as the standard. TCA/acetone precipitation of protein The supernatant of total proteins was precipitated overnight with 3 volumes of ice-cold (-20°C) 10% TCA/acetone containing 20 mM DTT 2. After centrifugation at 16,000 g for 30 min at 4°C, the supernatant was discarded, and the pellet was rinsed three times in ice-cold acetone containing 20 mM DTT. The pellet was air-dried, resuspended in rehydration buffer (6 M urea, 2 M thiourea, 0.5% Triton X-100, 0.5% IPG buffer, 20 mM DTT, 0.002% bromophenol blue) at 4°C overnight. IPG buffer was supplied by Amersham Biosciences. Protein content was determined by the 2-D Quant Kit (Amersham Biosciences) according to the manufacturer’s instruction. Two-dimensional electrophoresis The first dimension (isoelectric focusing) was performed on the IPGphor II (Amersham Biosciences) at 20°C with current limit at 50 μA per strip. Sample was dissolved in rehydration buffer as above and applied on IPG strip in the strip holder following the instruction from Amersham Biosciences. For 13-cm IPG strip (pH 4-7, Immobiline DryStrip, Amersham Biosciences), each strip was rehydrated at 30 V for 12 h, and then focused according to the preset program: 200 V (1.5 h), 500 V (1 h), 1,000 V (1 h), 4,000 V (1 h), 8,000 V (2 h), until the total Vh reached 19,960. For the 7-cm IPG strip (pH 4-7 or 3-10), each

strip was rehydrated at 30 V for 12 h, and then followed the program: 500 V (0.5 h), 1,000 V (0.5 h), 4,000 V (0.5 h), 5,000 V (2 h), until the total Vh reached 13,110. After the isoelectric focusing, strip was removed and equilibrated in 5 ml equilibration buffer (50 mM Tris, pH 8.8, containing 6 M urea, 30% glycerol, 2% SDS, 0.01% bromophenol blue) containing 1% (w/v) DTT for 15 min, and then transferred to 5 ml equilibration buffer containing 2.5% (w/v) iodoacetamide for another 15 min. The equilibrated strip was applied on the top of the 12.5% SDS-PAGE gel, sealed with 0.5% agarose, and then run the second dimension at 200 V for 4 h with Ruby (Hoeffer). Protein transfer and immunostaining Both one- and two-dimensional gels were transferred to PVDF membrane (Millipore) for 1.5 h at 400 mA using Transphor TE-62 (Hoeffer). The membrane was then blocked overnight with blocking solution (50 mM Tris, 0.25% gelatin, 150 mM NaCl, 5 mM EDTA, 0.05% Tween 20). The first antibody with appropriate dilution (e.g. 1:1,000, in blocking solution) was added to the membrane and incubated at room temperature for 2 h. The membrane was washed three times in PBST buffer (10 mM NaH2PO4, 130 mM NaCl, 0.05% Tween 20), and then probed with the second Ab (goat anti-mouse IgG and horseradish peroxidase conjugate, 1:10,000) for 1 h. After washing three times with PBST buffer, the enzyme activity on the blot was visualized through chemiluminesence by adding ECL Western Blotting Reagents (Pierce) according to the manufacturer’s instruction. Immunization Balb/c mouse (6-week old) was immunized (i.p.) with 1-ml emulsion containing 50 μg protein by mixing with equal volume of the synthetic adjuvant TiterMax. Booster shoots followed every two weeks with the same antigen dose for three months. The immune response was monitored by 2DE immunostaining with antisera obtained through the mouse tail-vein. Cell fusion Cell fusion was performed following the conventional protocol which was modified in our laboratory 3. One week before cell fusion, the mouse was injected (i.v.) with 50 μg soluble antigen without adjuvant. The spleen was

harvested, and all splenocytes (about 10 x 108 cells) were mixed with mouse myeloma (NS-1 or Sp2/0, 10 x 107 cells). The cell mixture was collected by centrifugation and resuspended well in a final volume of 1 ml in RPMI 1640 at 37°C. If Sp2/0 was used for fusion, the RPMI 1640 was replaced by DMEM (Dulbecco's modified Eagle's medium, GIBCO) throughout the procedure. They were then fused by mixing gently with 1 ml polyethyleneglycol (PEG 1500, Roche) in a period of 2 min. The PEG was diluted slowly by adding 10 ml RPMI 1640 (or DMEM) in 4 min, and then removed by centrifugation. The cell pellet was gently resuspended in 30 ml culture medium (RPMI 1640 or DMEM with 10% fetal calf serum) containing HAT (hypoxanthine, aminopterin, thymidine). The fused cells were distributed uniformly into three 96-well microtiter plates which were left undisturbed for 3 d in the cell incubator. Then one drop of RPMI 1640 containing hypoxanthine, thymidine and 10% fetal calf serum was added to each well on the plate every two or three days. The day of fusion was denoted as Day zero (D0). Hybridoma screening and identification On D14, supernatant (50 μL) from each well was sampled and analyzed by one-dimensional electrophoresis (1DE, 12.5% SDS-PAGE) and immunostaining (First screening). Positive clones were picked and expanded to T-25 culture; the one-week medium was used to reveal the binding pattern of mAb by 2DE immunostaining (First identification). Candidate clones were selected according to the 2DE blots, and were further subcloned by limiting dilution (one 96-well plate per candidate clone). Positive clones on the subcloned plate were screened again by 1DE and immunostaining (Second screening), and then transferred to T-25 for another 2DE analysis (Second identification). Duplicate clones were identified and eliminated by examining their 2DE immunostaining patterns during the identification steps. The selected clones were expanded to T-80 flask for producing mAb, and cell preservation. Antibody purification The cell supernatant was collected, and the immunoglobulin was fractionated by ammonium sulfate precipitation at 40% saturation. The precipitate was dissolved in 50 mM NaHCO3 buffer (pH 8.3), and then dialyzed at 4°C overnight. The precipitate formed during dialysis was removed by centrifugation before application to DEAE-Sephacel column. The column was washed with NaHCO3 buffer (as above), and then antibody fraction was eluted

with a linear gradient of NaCl (25 to 250 mM in NaHCO3 buffer, total 150 ml), which appeared at a concentration of approximately 200 mM NaCl. Preparation of affinity adsorbent Antibody prepared as above was dialyzed in coupling buffer (100 mM NaHCO3, 0.5 M NaCl, pH 8.3), and then coupled to the cyanogen bromide-activated Sepharose 4B (Amersham Biosciences) at the concentration of 5 mg antibody per ml gel. The reaction suspension was rotated end-over-end at 4°C. After overnight reaction, gel was washed twice with coupling buffer, and then incubated in 200 mM glycine (pH 8.0) for 2 h to block the remaining active groups. Totally, we have prepared 160 individual affinity adsorbents for every mAb in the First-stage bank. Affinity adsorption of the antigens Half milliliter of the affinity adsorbent from each mAb was pooled together for packing an affinity column (i.d. 2.5 x 16 cm) which was then equilibrated in coupling buffer. The total soluble proteins extracted from the bamboo shoots (?? mg) was slowly applied to this column, and the eluent was collected and applied back to the column. This process was repeated eight times to remove the antigens in the total proteins. The remaining proteins which could not be adsorbed by the immobilized antibodies was collected, and used for the antigen in the Second-stage immunization and mAb production.

Reference List 1. Gegenheimer,P. Preparation of extracts from plants. Methods Enzymol. 182, 174-193 (1990). 2. Gallardo,K. et al. Proteomics of Arabidopsis seed germination. A comparative study of wild-type and gibberellin-deficient seeds. Plant Physiol 129, 823-837 (2002). 3. Chen,H.M. et al. Regulation of the catalytic behaviour of L-form starch phosphorylase from sweet potato roots by proteolysis. Physiol Plant 114, 506-515 (2002).

Figure Legends Supplementary Figure 1 The immune response of the host animal (Mouse B) to complex antigens. (a) The total water soluble proteins (50 μg) from the rapid-growing bamboo shoots were analyzed by 2DE (13 cm IPG strip, pH 4-7) and stained with ammoniacal silver. (b-f) Pre-immune serum (b) and the antisera (c-f) were used to stain the 2DE blots of the total proteins (250 μg). Supplementary Figure 2 The final library of 2DE blots stained by individual mAb from the bank. (a-d) The First-stage bank amounts to 160 distinctive mAb. (e) The Second-stage bank has 32 distinctive mAb.

Supplementary Figure 1

Total protein

Pre-immune serum

4-week serum

a

b

c

d

e

f

6-week serum

8-week serum

12-week serum

Supplementary Figure 2 (a) 2DE library of First-stage mAb (Group 1, pH 4-7)

AB10A3

AB10A7

AB10A10

AB10A11

AB10B3

AB10B4

AB10B9

AB10B12

AB10C6

AD6D5

AE6A2

AE6A5

AE6A6

AE6A7

AE6A12

AE6B4

AE6B7

AE6B12

AE6C4

AE6C10

AE6E5

AE6E6

AE6F2

AE6F3

AE6F8

AE6F10

BF9G1

BF9G2

BF9G5

BF9H3

BE5A17

BF9F12

BF9E2

BF9E4

BG9B5

SA10D5

SA10E5

SC4E1

SK4A8

SF6D6

(b) 2DE library of First-stage mAb (Group 2, pH 3-10)

AB10A8

AB10A9

AB10A11

AB10A12

AB10B2

AD6A4

AD6C7

AD6G10

AE6A9

AE6B3

AE6B8

AE6C3

AE6C6

AE6C9

AE6D3

AE6D8

AE6D10

AE6E9

AE6E12

AE6F6

BE5A2

BE5A3

BE5A5

BE5A9

BE5A11

BE5C9

BF9E5

BF9E8

BF9F6

BF9F5

BF9F8

BF9G3

BF9G4

BF9G6

BF9G9

BF9G12

BF9H4

BG9A6

BG9A10

BG9B7

(c) 2DE library of First-stage mAb (Group 3, pH 3-10)

BH10A6

BH10C3

CE2F8

SA4C1

SA4C5

SA4C6

SA4E1

SA7D1

SA7D2

SA7E6

SA8H3

SA8H4

SA10C1

SA10C3

SA10C6

SA10D2

SA10D8

SA10D10

SA10E3

SA10E7

SA10E10

SB2A3

SB2A4

SB2A10

SB2B3

SB2B4

SB2B6

SB2C1

SB2C8

SB2D1

SB2D6

SB2E1

SB2G1

SB2G6

SB2H1

SB2H2

SB2H3

SB2H4

SB2H6

SB2H9

(d)

2DE library of First-stage mAb (Group 4, pH 3-10)

SB4A1

SB4A3

SC3B2

SC3G8

SC3E1

SC4E1

SC4E2

SC4E3

SC4E4

SC4E5

SC4E6

SC9A12

SC9B11

SF6A12

SF6D1

SF6D2

SF6D7

SF10A1

SP1C5

SP1B2

SP1B6

SK1A3

SK1A4

SK1A10

SK3D7

SK5G2

SK5G6

SK5H4

SK5H10

SK6H2

SK9C2

SK9C4

SK9C7

SK9C10

SK9D2

SK9E2

SK9H3

SL3G5

SL5B8

SPL7E5

(e) 2DE library of Second-stage mAb (Group 5, pH 4-7)

2SM1A3

2SM5A3

2SM5A5

2SM3A7

2SN1A2

2SP3A2

2SP3A5

2SP5A5

2SP5A10

2SP11A10

2SP3A8

2SP3B3

2SP5A1

2SQ2A10

2SQ12A4

2SQ12A12

2SR1A1

2SR1A13

2ST7C6

2ST7C7

2SU2A1

2SU2A7

2SU6A8

2SU6B9

2SU12A4

2SU12A3

2SV5A1

2SV5A8

2SV5A9

2SV5B8

2SV5B10

2SW12B1