Antibody fusion proteins with human ribonucleases 1 to 8

Abstract

ImmunoRNases combine tumor targeting by antibodies with the cytotoxic action of ribonucleases from the RNase A superfamily. This study investigated for the first time all catalytic active human RNase A family members (1 to 8) as effector components of antibody fusion proteins. ImmunoRNase fusion proteins were constructed using the CD30-specific bivalent recombinant scFv-Fc antibody SH313-B5. Production of the resulting entirely human immunoRNases 1 to 8 was done in mammalian cells by secretion of active forms. The immunoRNases mediated CD30-specific cell binding and showed ribonucleolytic activity. Interestingly, immunoRNases 1 and 2 were active in the presence of up to 5-/20-fold molar excess of the pancreatic RNase inhibitor (RI), which is supposed to efficiently inhibit all human RNase A activity. ImmunoRNases 3, 4, 6 and 7 were only inhibited by several fold molar excess of RI, whereas immunoRNases 5 and 8 were already completely inactive at equimolar RI concentrations. Compared to free RNases, activity and RI sensitivity were not significantly changed by antibody fusion or dimerisation. ImmunoRNase3 and 5 mediated tumor growth inhibition at low nanomolar concentrations. Anti-tumor activity was antigen-specific and did not show any correlation with ribonucleolytic activity or RI sensitivity.

Keywords

ImmunoRNase human RNase A superfamily RNases 1 to 8 human antibody CD30 lymphoma antibody-based therapy

1. Introduction

In the last two decades, monoclonal antibodies (mAbs) became one of the fastest growing class of biological therapeutics on the market [61]. However, the success of unconjugated mAbs for cancer therapy has lacked behind initial hopes in respect of clinical efficacy. Therefore, antibodies have been intensively studied as vehicle to deliver highly cytotoxic payloads into tumors, but only recently, the first targeted toxins have been approved for therapy. The first respective drug, Mylotarg (Gentuzumab ozogamicin) was withdrawn from the market after no improvement in clinical benefit was observed in a post-approval study. The addition of a cytotocic payload (microtubule inhibitor DM1) improved the efficacy of trastuzumab, and Brentuximab vedotin is approved for the treatments of lymphomas. Key factors for such a reagent are efficient targeting and intracellular uptake, which are both depending on the antigen. Brentuximab, which recognises human CD30, has demonstrated that CD30 target, also referred to as Ki-1 [14, 16], is a suitable target to study human “immunotoxin”-like antibody fusion proteins. While unconjugated anti-CD30 mAbs achieved disappointing results in clinical trials of relapsed and refractory HL patients [17, 18, 59], The antibody drug conjugate (ADC) brentuximab vedotin was clinical effective, displayed manageable toxicities and was approved for treatment of CD30 ${}^{+}$ relapsed and refractory HL and ALCL in 2012 [59]. CD30 is over-expressed on most Hodgkin lymphomas (HL) and some Non-Hodgkin lymphomas (NHL), particularly anaplastic large cell lymphoma (ALCL) [5, 26, 77], but shows only limited expression on non-tumor tissues like activated proliferating or virus-infected lymphocytes, in decidual and endometrial cells and in lymph node cells in the parafollicular area [14, 26]. Therefore, CD30 represents a suitable tumor-associated antigen providing tumor specificity with low risk of peripheral toxicity [36]. CD30 belongs to the tumor necrosis factor receptor superfamily [14] and activates different signal pathways causing pleiotropic effects like differentiation, activation, proliferation or apoptosis depending on cell type, differentiation stage and additional environmental stimuli [14, 18, 26, 28, 77]. Although CD30 ${}^{+}$ HL and ALCL are well treated by chemotherapy and radiation [13, 34, 67], first line standard therapy fails in about 10–15% of the cases [17, 30, 34, 44]. Retreatment of these patients frequently results in relapse with very poor prognosis [18, 30].

Table 1
In this study used oligonucleotides

Oligonucleotide name	#	5’-3’ sequence
TS_RNase2_f	720	TTCTCACAGGAGCTACAGC
TS_RNase2_r	721	TGAGTGATGATGAGGAGTGC
TS_RNase3_f	722	GGAGCCACAGCTCAGAGACT
TS_RNase3_r	723	GATGAGGACTGCTGATACAGG
TS_RNase7_f	724	ACGAAGACCAAGCGCAAAGC
TS_RNase7_r	725	AGGTCAGCCAAAGAGCAAGC
XWe_RNase8_if	824	CATCCACCTACCTCCTCAC
XWe_RNase8_ir	825	AGCAGCAGCATACAGTCTG
XWe_RNase1_f	826	GTCAGAAGATCTATGAAGGAATCCCGGGCCAA
XWe_RNase1_r	827	CGGCCGCTCTAGATTTAGGTAGAGTCCTCCACAGAAG
XWe_RNase2_f1	848	GTCAGAAGATCTATGTCACTCCATGTCAAACCTC
XWe_RNase2_r	849	CGGCCGCTCTAGATTTAGATGATTCTATCCAGGT
XWe_RNase3_f	850	GTCAGAAGATCTATGAGACCCCCACAGTTTACG
XWe_RNase3_r	851	CGGCCGCTCTAGATTTAGATGGTGGTATCCAG
XWe_RNase4_f	852	GTCAGAAGATCTATGCAGGATGGCATGTACCAG
XWe_RNase4_r	853	CGGCCGCTCTAGATCTAACCGTCAAAGTGCAC
XWe_RNase5_f	854	GTCAGAAGATCTATGCAGGATAACTCCAGGTAC
XWe_RNase5_r	855	CGGCCGCTCTAGATTTACGGACGACGGAAAATTG
XWe_RNase6_f	856	GTCAGAAGATCTATGTGGCCTAAGCGTCTCAC
XWe_RNase6_r	857	CGGCCGCTCTAGATTTAGAGAATACTATCTAAGTG
XWe_RNase7_f	858	GTCAGAAGATCTATGAAGCCCAAGGGCATGAC
XWe_RNase7_r	859	CGGCCGCTCTAGATCTAAAGGACTCTGTCCAAG
XWe_RNase8_f	860	GTCAGAAGATCTATGAAGCCCAAGGACATGAC
XWe_RNase8_r	861	CGGCCGCTCTAGATTTAGACAACTTTATCCAAG
XWe_SpeI_RNase2_f	1118	GGTAGTACTAGTTCACTCCATGTCAAACCTCC
XWe_SpeI_RNase3_f	1119	GGTAGTACTAGTAGACCCCCACAGTTTACGAG
XWe_SpeI_RNase4_f	1120	GGTAGTACTAGTCAGGATGGCATGTACCAG
XWe_SpeI_RNase5_f	1121	GGTAGTACTAGTCAGGATAACTCCAGGTAC
XWe_SpeI_RNase6_f	1122	GGTAGTACTAGTTGGCCTAAGCGTCTCACC
XWe_SpeI_RNase7_f	1123	GGTAGTACTAGTAAGCCCAAGGGCATGACCTC
XWe_SpeI_RNase8_f	1124	GGTAGTACTAGTAAGCCCAAGGACATGACATC

Alternatively to ADCs, immunotoxins made of of mAbs conjugated or fused to bacterial or plant toxins like diphtheria toxin, Pseudomonas exotoxin A, saporin or ricin have been studied for more than two decades [19, 36, 43, 58, 61]. However, they frequently suffered from high unspecific toxicity and immunogenicity [5, 23, 58]. To overcome toxicity and immunogenicity issues, a number of mammalian and human enzymes have been investigated to replace heterologous toxin component [40, 53, 55, 56, 66, 72, 73]. Particularly, vertebrate RNase A family enzymes have been considered as alternative effector moiety. Human RNase A members are present in nearly all organs, tissues and body fluids which should reduce the risk of non-specific toxicity and immunogenicity [58]. Even heterologous bovine RNase A or ranpirnase (Onconase ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}\textregistered}$ ) from the leopard frog Rana pipiens were well tolerated in human patients [1, 38]. Once RNases are delivered to the cytosol of cells, they can degrade RNA substrates and can cause cytotoxic effects comparable to highly potent toxins like ricin [36, 58].

Therefore, RNase A proteins were tested in conjugation with mAbs or fusion proteins, which are referred to as immunoRNases [54]. The amphibian Onconase, human RNases 1, 2 or 5 in combination with mAbs achieved strongly increased cytotoxicity and demonstrated the therapeutic potential of immunoRNases [10, 36, 58, 80]. The proposed cytotoxic mechanism of immunoRNases requires internalization and release of the active RNase moiety into the cytosol, where it degrades RNA substrates and inhibits protein synthesis [11, 58]. CD30 specific immunoRNase fusion proteins employing human pancreatic RNase (RNase1) or angiogenin (RNase5) inhibited CD30 ${}^{+}$ lymphoma cell lines in vitro and in vivo [7, 37]. Contrarily, RNase1 fused to different tumor-specific immunoglobulin (Ig)G in therapeutic pipeline did not achieve efficient antitumor effects, which was not improved by RI evading RNase variants or by introduction of different putatively cleavable linker sequences [57].

In this study all catalytic active human RNase A members (RNases 1 to 8) were compared for their effects when fused to the described CD30-specific human bivalent antibody SH313-B5 [78]. The resulting immunoRNases 1 to 8 are entirely of human origin and were tested for antigen and cell binding, anti-tumor effect, ribonucleolytic activity and the sensitivity against the cytosolic human RNase-inhibitor (RI). RI is ubiquitously expressed in mammalian cells and binds RNase A with an extremely high affinity [46, 58]. It was previously proposed that the RI could abolish the anti-tumor effect of human immunoRNases by inhibiting catalytic activity of all human RNase A proteins [12, 35, 42, 46]. Thus, this study gives the first comprehensive overview about immunoRNases employing all catalytic active members of human RNase A superfamily including ribonucleolytic activity, RI sensitivity and potential anti-tumor properties.

2. Results

2.1 Production and purification of immunoRNases

Sixteen different immunoRNases were constructed by fusing one of the human RNases 1 to 8 to the carboxy terminus either of the CD30-specific bivalent human scFv-Fc antibody SH313-B5 or the human negative control antibody MS112-IIB1. All immunoRNases were produced in mammalian cells by secretion in soluble form and purified by protein A affinity chromatography resulting in volumetric yields between 3 mg/L and 60 mg/L (Table 3). SDS-PAGE analysis revealed high purity of all immunoRNases (Fig. 1). The CD30-specific immunoRNases corresponded to a molecular weight of 70–75 kDa except for immunoRNases 2 and 3, which migrated at 80–85 kDa. Moreover, immunoRNases 1, 2, 3, 7, and 8 showed blurry protein bands suggesting heterogeneous glycosylation. ImmunoRNase7 displayed some proteolysis, as a faint additional band at $\sim$ 60 kDa appeared which corresponds to the molecular mass of the scFv-Fc antibody fragment.

2.2 Production and purification of free human RNases

Free human RNases 1 to 8 without antibody moiety were produced for reference in respect of catalytic activity and RI inhibition. The expression of free RNases in mammalian cells resulted only in very low yields of recombinant protein (data not shown). However, all human RNases 1 to 8 could be successfully produced as Fc-TEV ${}_{cs}$ -RNase fusion proteins in mammalian cells and purified by protein A affinity chromatography. Then, the carboxy terminal RNase moieties were cleaved from the Fc-part with TEV protease. The TEV protease cleavage site ENLYFQ G should leave only one additional glycine residue at the amino-terminus of the RNases compared to their natural mature sequence, which should not interfere with ribonucleolytic activity and RI binding [8, 9, 15, 22, 62]. Although TEV protease digestions were generally successful for all Fc-TEV ${}_{cs}$ -RNase1 to 8 fusion proteins, only RNases 1, 4, 5 and 7 were efficiently cleaved from the fusion protein (Fig. 2A) to allow subsequent purification by SEC (Fig. 2B). In contrast, cleavage of Fc-TEV ${}_{cs}$ -RNase7 was not complete, which can be seen by the non-cleaved fusion protein band at $\sim$ 50 kDa (Fig. 2A). TEV digestion of Fc-TEV ${}_{cs}$ -RNases 2, 3, 6 and 8 was either not site-specific or associated with RNase degradation (Fig. 2A). Unspecific cleavage or degradation was not associated with the Fc-part, which was detected at $\sim$ 30 kDa for all Fc-TEV ${}_{cs}$ -RNase1 to 8 fusion proteins in SDS-PAGE confirmed by Western immunoblot with a Fc-specific antibody conjugate (data not shown). Larger amounts of TEV protease and/or longer incubation time did not result in more RNase-product of Fc-TEV ${}_{cs}$ -RNases 2, 3, 6 and 8 (data not shown). Nevertheless, this method allowed the SEC purification of RNases 1, 4, 5 and 7 with high purity and with relatively high volumetric yields (Table 3). Interestingly, RNase1 purification resulted in two different molecular weight fractions, which were both proved to be RNase1 using Western immunoblot with a rabbit anti-RNase1 serum, but seem to differ in their degree of glycosylation (supplement Fig. 1). The RNase1 protein fraction with lower molecular weight ( $\sim$ 17 kDa) was used for further studies. RNases 4, 5 and 7 displayed single protein bands in SDS-PAGE after purification (Fig. 2B).

Table 2
Molecular masses and glycosylation of human RNases

Human RNase	Molecular mass [kDa]	Glycosylation ${}^{*}$
1	15.0	$+$
2	16.3	$+$
3	16.0	$+$
4	14.3	$-$
5	14.6	$-$
6	15.1	$+$
7	15.0	$+$
8	14.7	$-$

${}^{*}$ UniProt (www.uniprot.org).

Figure 1.

SDS-PAGE analysis of immunoRNases 1 to 8. A total of 1 $\mu$ g of purified immunoRNases 1 to 8 (lanes 1 to 8) were analyzed by SDS-PAGE and coomassie staining. M: protein molecular mass marker; see table 2 for molecular masses and glycosylation of RNases.

Table 3

Volumetric yields of immunoRNases 1 to 8 and free RNases 1, 4, 5 and 7

Protein	Antibody	Yield
	specificity	[mg/L]
ImmunoRNase1 ${}^{*}$	CD30	60.0
ImmunoRNase1 ${}^{*}$	Control	32.0
Free RNase1 ${}^{**}$	–	5.40
ImmunoRNase2 ${}^{*}$	CD30	28.0
ImmunoRNase2 ${}^{*}$	Control	22.7
ImmunoRNase3 ${}^{*}$	CD30	10.0
ImmunoRNase3 ${}^{*}$	Control	8.0
ImmunoRNase4 ${}^{*}$	CD30	23.5
ImmunoRNase4 ${}^{*}$	Control	2.7
Free RNase4 ${}^{**}$	–	0.6
ImmunoRNase5 ${}^{*}$	CD30	34.1
ImmunoRNase5 ${}^{*}$	Control	5.3
Free RNase5 ${}^{**}$	–	4.0
ImmunoRNase6 ${}^{*}$	CD30	28.8
ImmunoRNase6 ${}^{*}$	Control	6.4
ImmunoRNase7 ${}^{*}$	CD30	7.5
ImmunoRNase7 ${}^{*}$	Control	21.3
Free RNase7 ${}^{**}$	–	14.6
ImmunoRNase8 ${}^{*}$	CD30	19.2
ImmunoRNase8 ${}^{*}$	Control	16.5

${}^{*}$ After Protein A affinitiy chromatography. ${}^{**}$ After Protein A affinity chromatography and TEV-protease cleavage followed by size exclusion chromatography.

Figure 2.

Production and purification of free recombinant human RNase 1 to 8. The human RNases 1 to 8 were produced as Fc-TEV ${}_{cs}$ -RNase fusion proteins in HEK293-6E cells, purified by protein A affinity chromatography and cleaved by incubation with TEV protease. A total of 3 $\mu$ g of the cleaved Fc-TEV ${}_{cs}$ -RNase 1 to 8 proteins (lane 1 to 8) was analyzed by SDS-PAGE and coomassie staining. Expected protein fragments are indicated (A). The free recombinant RNase fragments were obtained by SEC. A total of 1 $\mu$ g of RNase 1, 4, 5, and 7 protein fragment (lane 1, 4, 5, and 7) was analyzed by SDS-PAGE and coomassie staining (B). M: protein molecular mass marker.

2.3 RNase activity

Ribonucleolytic activity was analyzed using two different substrates (rU, rC), which both contained only a single RNase cleavage site after C and U, respectively. Cleavage of the substrate resulted in a time-dependent increase of fluorescence as shown for RNase1 (Supplement 2). The negative control without RNase did not reveal any increase of fluorescence, which excludes potential contaminations of other RNases during production, purification and the activity assays. Enzyme efficiencies ( $k_{cat}$ / $K_{M}$ ) are summarized in Table 4. The enzyme efficiencies of the immunoRNases 1, 4, 5 and 7 were similar or only slightly lower than the values measured for the corresponding free recombinant RNases 1, 4, 5 and 7.

Table 4
Enzyme efficiencies of immunoRNases 1 to 8 and of free RNases 1, 4, 5 and 7

		Enzyme efficiency
	Antibody	( $k_{cat}$ / $K_{M}$ ) [M ${}^{-1}$ s ${}^{-1}$ ]
Protein	specificity	rU ${}^{*}$	rC ${}^{*}$
ImmunoRNase1 ${}^{*}$	CD30	3 $\times$ 10 ${}^{6}$	3 $\times$ 10 ${}^{6}$
ImmunoRNase1 ${}^{*}$	Control	5 $\times$ 10 ${}^{6}$	8 $\times$ 10 ${}^{6}$
Free RNase1 ${}^{**}$	–	2 $\times$ 10 ${}^{7}$	2 $\times$ 10 ${}^{7}$
ImmunoRNase2 ${}^{*}$	CD30	1 $\times$ 10 ${}^{6}$	1 $\times$ 10 ${}^{6}$
ImmunoRNase2 ${}^{*}$	Control	2 $\times$ 10 ${}^{6}$	2 $\times$ 10 ${}^{6}$
ImmunoRNase3 ${}^{*}$	CD30	5 $\times$ 10 ${}^{4}$	5 $\times$ 10 ${}^{4}$
ImmunoRNase3 ${}^{*}$	Control	3 $\times$ 10 ${}^{5}$	2 $\times$ 10 ${}^{5}$
ImmunoRNase4 ${}^{*}$	CD30	7 $\times$ 10 ${}^{5}$	5 $\times$ 10 ${}^{5}$
ImmunoRNase4 ${}^{*}$	Control	1 $\times$ 10 ${}^{6}$	9 $\times$ 10 ${}^{5}$
Free RNase4 ${}^{**}$	–	2 $\times$ 10 ${}^{6}$	9 $\times$ 10 ${}^{5}$
ImmunoRNase5 ${}^{*}$	CD30	2 $\times$ 10 ${}^{4}$	2 $\times$ 10 ${}^{4}$
ImmunoRNase5 ${}^{*}$	Control	4 $\times$ 10 ${}^{4}$	2 $\times$ 10 ${}^{4}$
Free RNase5 ${}^{**}$	–	9 $\times$ 10 ${}^{4}$	4 $\times$ 10 ${}^{4}$
ImmunoRNase6 ${}^{*}$	CD30	4 $\times$ 10 ${}^{4}$	4 $\times$ 10 ${}^{4}$
ImmunoRNase6 ${}^{*}$	Control	2 $\times$ 10 ${}^{5}$	1 $\times$ 10 ${}^{5}$
ImmunoRNase7 ${}^{*}$	CD30	4 $\times$ 10 ${}^{4}$	2 $\times$ 10 ${}^{5}$
ImmunoRNase7 ${}^{*}$	Control	6 $\times$ 10 ${}^{5}$	7 $\times$ 10 ${}^{5}$
Free RNase7 ${}^{**}$	–	4 $\times$ 10 ${}^{4}$	2 $\times$ 10 ${}^{5}$
ImmunoRNase8 ${}^{*}$	CD30	2 $\times$ 10 ${}^{4}$	2 $\times$ 10 ${}^{4}$
ImmunoRNase8 ${}^{*}$	Control	3 $\times$ 10 ${}^{4}$	2 $\times$ 10 ${}^{4}$

${}^{*}$ Abbreviations: rU: 6’-FAM-dA-rU-(dA) ${}_{2}$ -6’-BHQ1 and rC: 6’-FAM-dA-rC-(dA) ${}_{2}$ -6’-BHQ1 with dA, desoxy-adenine; rC, cytidine; rU, uridine; 6’-FAM, 6-carboxyfluorescein; 6’-BHQ1, Black-Hole-Quencher1. ${}^{**}$ Free recombinant RNases were produced as Fc-TEV ${}_{CS}$ -RNase fusion proteins cleaved by TEV protease.

The highest in vitro ribonucleolytic activity was measured for free human RNase1, while the activities of the immunoRNases1 were $\sim$ 10-fold lower. These data corresponded well to values previously measured by other groups [4, 47, 65] and are in the same range like the activity of bovine RNase A [31]. Different to previous data [4, 65], there was no preference for the cleavage 3’ after U or C using our substrates. ImmunoRNases2 confirmed slightly lower catalytic efficiency to both substrates compared to immunoRNases1 [64]. Human RNase3 was described to have a lower activity than RNase2 [64], which was also seen in this study for the immunoRNases. The activity of immunoRNases4 and recombinant free RNase4 were in the same range as previously shown for RNase4 with another U-containing RNA substrate, whereas the preference for cleavage 3’ after U over C was less pronounced [24]. Other groups even described a stronger preference for U over C cleavage [24, 68, 71]. Activities of the immunoRNases5 and free recombinant RNase5 were in the same order of magnitude for all analyses. They mediated the lowest activity among all tested immunoRNases and free RNases, which is in accordance with the results from other research groups [64, 70]. However, their enzymatic efficiencies were $\sim$ 100-times higher in this study than previously described in any other study [31]. Moreover, an $\sim$ 2-fold substrate preference for U over C cleavage was observed, whereas other groups described a preference for C [25, 31]. RNase6 was described to have a 40-fold lower activity than RNase2 [64], which was approximately confirmed for the immunoRNases6 in this study. ImmunoRNases7 and recombinant free RNase7 showed a clear substrate preference for C over U, but enzyme efficiency was lower than previously described for RNase7 isolated from human epidermis [21]. The low activity of immunoRNases8 confirmed previous data [64, 81] and was almost at the level of immunoRNases5 and free RNase5 in this study.

Figure 3.

RI mediated inhibition of immunoRNases 1 to 8 and free recombinant RNases 1, 4, 5 and 7. Ribonucleolytic activity was measured with rU substrate in the presence of up to 20-fold molar excess of RI. The concentration of active RNases was normalized to samples incubated without RI (100%). Average values and standard deviations from triplicate experiments are given. n.d.: not determined.

Figure 4.

Analytic SEC of immunoRNases 1 to 6 and 8. PBS was used as running buffer. ImmunoRNase 7 was not analyzed due to instability of the fusion protein.

2.4 Inhibition of RNase activity by human RNase inhibitor (RI)

The inhibition of the ribonucleolytic activity of the CD30-specific immunoRNases 1 to 8 and the free recombinant human RNases 1, 4, 5 and 7 was analyzed in the presence of the human cytosolic RI, which was added in up to 20-fold molar excess 15 minutes before the substrate (Fig. 3). The RI had very different inhibitory effect on the catalytic activity of the tested free RNases 1, 4, 5 and 7 and the immunoRNases 1 to 8, which can be arranged into three groups. The first group is not or only slightly inhibited at a 5- to 20-fold molar excess of RI like immunoRNase1 and the corresponding free RNase1. ImmunoRNase 2 also belongs into this group, but may show higher RI sensitivity than RNase1 at higher excess of the inhibitor. ImmunoRNases 3, 4, 6 and 7 as well as the recombinant free RNases 4 and 7 belong to the second group, which is not completely inhibited at equimolar concentration of the RI, but the ribonucleolytic activity is strongly reduced in the presence of 5- to 20-fold molar excess of the RI. Both, immunoRNase4 and free RNase4 were completely inhibited at 20-fold molar excess of RI, whereas immunoRNase 3, 6 and 7 still had significant (23–29%) residual ribonucleolytic activity. The third group comprising immunoRNases 5 and 8 and free RNase5 was completely inhibited in the presence of equimolar concentrations of RI. Titration of RNase5 revealed a highly efficient strong 1:1 interaction with RI [32].

RI-mediated inhibition of immunoRNases 1, 4, 5 and 7 was similar compared to the corresponding free RNases 1, 4, 5 and 7, respectively. Therefore, the fusion of RNases and antibody did not affect the RI-mediated inhibition of these immunoRNases. Moreover, the Fc-mediated dimerization did not have any major influence on the RI-mediated inhibition.

2.5 SEC analysis of immunoRNases

The CD30 specific immunoRNases were analyzed by SEC (Fig. 4) and revealed one major peak except for immunoRNase7, which showed a pattern of degradation (data not shown). Homodimeric proteins were expected due to the dimerization of the Fc-part of the antibody moiety. The analyzed immunoRNases showed two generally different distributions. ImmunoRNases 4 and 5 had their major peak at about 130–150 kDa, which corresponded to a homodimeric, unglycosylated scFv-Fc-RNase form. In contrast, immunoRNases 1, 2, 3, 6 and 8 migrated at higher molar masses, indicating substantial glycosylation.

Figure 5.

Cell binding of immunoRNases 1 to 8. Purified immunoRNases 1 to 8 were tested for binding to CD30 ${}^{+}$ Karpas299 cells and CD30 ${}^{-}$ Jurkat cells by flow cytometry. The results are shown as ratio of the median fluorescence intensities (MFI) compared to control staining with the FITC conjugated secondary antibody conjugate (2 ${}^{\rm nd}$ ab. $=$ 1). The antibody SH313-B5 without RNase was used as additional control (+).

2.6 Binding of immunoRNases to CD30

{}^{+}

lymphoma cells

The binding of the immunoRNases to CD30 ${}^{+}$ Karpas299 cells was analyzed by immunofluorescence staining and flow cytometry (Fig. 5). The CD30-specific immunoRNases bound specifically to CD30 ${}^{+}$ Karpas299 cells except for immunoRNase7, which displayed binding to CD30 ${}^{-}$ Jurkat cells and also showed a 2–3-fold higher binding to CD30 ${}^{+}$ cells than the parental antibody SH313-B5 and the other immunoRNases, which was probably caused by the RNase7 moiety. ImmunoRNase5 bound with double intensity to CD30 ${}^{+}$ cells compared to the parental antibody and the other immunoRNases (except immunoRNase7), which could be due the expression of a putative receptor [27] on this cell line.

Figure 6.

Tumor growth inhibition by immunoRNases 1 to 8. A) CD30 specific immunoRNases 1 to 8 were incubated with CD30 ${}^{+}$ Karpas299 (left) and CD30 ${}^{-}$ Jurkat cells (right) for 3 days. Viable cells were analyzed using an MTT-assay. Both cell lines were also tested with the parental antibody SH313-B5 (+). B) Two variants of immunoRNase3 containing the CD30-specific antibody SH313-B5 or a control antibody, respectively, were tested for growth inhibition of both cell lines. The corresponding antibodies were tested for comparison.

2.7 Inhibition of tumor cell growth by immunoRNases

The anti-tumor effect of immunoRNases 1 to 8 was measured by an MTT assay using CD30 ${}^{+}$ Karpas299 lymphoma cells (Fig. 6A). The parental antibody SH313-B5 already mediated growth inhibition of CD30 ${}^{+}$ Karpas299 cells with a half inhibitory concentration (IC ${}_{50}$ ) of about 100 nM as previously shown [78]. The immunoRNases 3 and 5 increased growth inhibition of CD30 ${}^{+}$ Karpas299 cells compared to the parental antibody, whereas immunoRNases 1, 2, 4, 6, 7 and 8 only showed similar or even lower inhibition than the parental scFv-Fc antibody SH313-B5. Neither antibody SH313-B5 nor any of the immunoRNases 1 to 8 showed significant inhibitory activity towards CD30 ${}^{-}$ Jurkat cells. ImmunoRNase3 mediated the strongest anti-tumor effect followed by immunoRNase5 (Fig. 6A). ImmunoRNase3 and the parental antibody SH313-B5 were analyzed in more detail including additional negative controls like an antibody and immunoRNase3 variant, which were not binding to human cells (Fig. 6B). The immunoRNase3 specifically inhibited CD30 ${}^{+}$ Karpas299 cells with an IC ${}_{50}$ value of $\sim$ 1 nM, which is 100-fold more efficient than the parental antibody. CD30 ${}^{-}$ Jurkat cells were not harmed by the CD30-specific immunoRNase3.

3. Discussion

ImmunoRNases consisting of human antibodies and human RNase A superfamily members represent an alternative to ADCs or immunotoxins for treatment of tumors, because they promise to overcome low efficacy, liver or kidney toxicity as well as immunogenicity and off-target binding [5, 23, 58]. Since ribonucleolytic activity has been proposed to be of central importance for the cytotoxic function of immunoRNases [36, 54, 58], the catalytic active human RNases 1 to 8 are of particular interest for tumor therapy. Consequently, in this study the human RNases 1 to 8 were fused to the CD30-specific human IgG-equivalent scFv-Fc antibody SH313-B5 [78] for direct comparison. Yields varied depending on the employed RNase moiety, but were comparable to immunotoxin fusion proteins [29, 37, 58]. Significantly, all of the immunoRNases 1 to 8 were produced in active form and secreted to the supernatant by human embryonal kidney cells in concentrations in the mg/L range. The producing cells were not harmed, which confirms previous results suggesting very low unspecific cell toxicity obtained with another human immunoRNase1 variant [37].

In contrast to these immunoRNase fusion proteins, the recombinant production of free human RNases 1 to 8 in their mature form was found to be very inefficient in E. coli (data not shown), which may to be due to high content of disulfide bonds reducing the yield of functional protein. Mammalian production allows for functional expression including correct formation of the 3–4 disulfide bridges and glycosylation, which is described for RNases 1, 2, 3, 6 and 7 [3, 24, 48, 50, 51, 52, 64, 68, 70, 71, 75, 81, 82]. However, human RNases 1 to 8 were only inefficiently secreted from HEK293 cells, which may indicate cell-dependent secretory pathways. Such alternative secretion pathways are described, for example for the RNases 2 and 3, which are stored in specialized vesicle of eosinophiles and are released upon stimulation [49, 51, 75]. To enhance recombinant secretory production, the human RNases 1 to 8 were fused to the Fc-part, which can promote expression and secretion of other proteins in mammalian cell lines [9, 37]. RNases 1, 4, 5 and 7 were successfully purified after site-directed cleavage from the Fc-moiety using TEV protease [8, 9, 15, 22, 62], whereas RNases 2, 3, 6 and 8 were either sensitive to non-specific TEV cleavage or degradation.

All immunoRNases 1 to 8 and the free recombinant human RNases 1, 4, 5 and 7 exhibited ribonucleolytic activity. Free RNases 1, 4, 5 and 7 revealed similar or slightly higher enzyme efficiencies compared to the much larger dimeric immunoRNases, which can be due to their higher diffusion rate and better accessibility of their active center [2]. Other groups also described lower activity of immunoRNase fusion proteins, e.g. only 6–13% activity of an immunoRNase2 fusion protein compared to free RNase2 [41].

In this study, the immunoRNases and recombinant free RNases revealed activity corresponding to previously published data. However, there were differences regarding the preference to cleave after U or C compared to previous publications, which may be explained by differences of the employed expression systems, activity assays and test substrates. Moreover, complex substrates with potential secondary structures were not investigated in this study. Among all tested immunoRNases, only immunoRNase/RNase 5 showed a significantly higher activity than described by any other group [31], which we assume is caused by the mammalian expression.

Mammalian cells express high levels of cytosolic RI, which is supposed to bind to all members of the RNase A superfamily with extremely high affinities [35, 42, 46] and to inactivate their enzymatic activity efficiently to protect cells from RNA damage by entrance of RNase A proteins [64]. Lowering affinities between RNases and RI enhanced the RNase-mediated cytotoxicity [4, 6, 35, 69, 74, 79]. Fusion with an antibody moiety or insertion of additional domains into the RNase structure can also lower RI interaction and inhibition of RNase activity [4, 69]. In this study, the RI had very different effect on the enzymatic activity of the tested immunoRNases and RNases. Since no major differences were observed between immunoRNases 1, 4, 5 and 7 and the corresponding free RNases 1, 4, 5 and 7, the fusion with the antibody moiety as well as the dimerization mediated by the Fc-moiety does not seem to influence the inhibition by the RI.

Three major groups of immunoRNases (and RNases) were identified according to different inhibition characteristics in the presence of the RI. The highly efficient inhibition of immunoRNases 5 and 8 and the corresponding free RNase5 indicates a 1:1 inhibitory interaction with the cytosolic RI, which was previously described with inhibitory concentrations in the femtomolar range [24, 32, 64, 68]. In contrast, immunoRNases 3, 4, 6 and 7 as well as the free RNases 4 and 7 were only partially inhibited at equimolar concentrations of RI. ImmunoRNases 1 and 2 as well as free recombinant RNase1 were not or only slightly inhibited by the RI even at 5- to 20-fold molar excess of RI, which was quite unexpected [4, 6]. However, very recently, several IgG-based immunoRNases1 fusion proteins were described to be not inhibited by a 50-fold molar excess of RI in an anti-tumoral study [57], which may require a new view on RI:RNase1 interaction and its physiological importance. It should be noticed that all immunoRNases and free RNases were incubated with the RI 15 minutes before addition of the substrate to allow to reach the equilibrium of RI:RNase complex formation before starting the activity assay. It cannot be entirely excluded that previous studies differ from these results due to different substrates and RNase activity assays, because small hypersensitive fluorogenic substrates [31] may still access the active site after RI-binding. However, in this case different models would be required to explain the interaction of RI with individual RNase A family members.

The immunoRNases 1 to 8 formed stable disulfide-linked scFv-Fc-RNase homodimers. Only immunoRNase7 showed a tendency of degradation. Moreover, immunoRNases 1, 2, 3 and 6 showed a pattern of high glycosylation, which was dependent on the RNase moiety, because all fusion proteins shared the same scFv-Fc antibody moiety. Glycosylation of RNases 1, 2, 3, 6 and 7 is in accordance with previous investigations [20, 63, 64]. ImmunoRNases 4, 5 and 8 did not show any signs of high or diverse glycosylation, which was confirmed for the corresponding free RNases 4, 5 and 8 by other groups [20, 64].

Each of the immunoRNases 1 to 8 containing the anti-CD30 antibody SH313-B5 specifically bound to CD30 ${}^{+}$ lymphoma cells but not to CD30 ${}^{-}$ cells, with the exception of immunoRNase7, while it remains unclear if this binding was unspecific or due to putative RNase7 receptor activity on this particular cell line. Moreover, binding of immunoRNase7 to CD30 ${}^{+}$ cells was 2–3 fold stronger compared to the other immunoRNases and the parental antibody, which may be a hint for an avidity effect of combined antibody- and RNase-mediated cell binding. ImmunoRNase5 too showed 2-fold stronger binding to CD30 ${}^{+}$ Karpas299 cells than the parental antibody, but no binding to CD30 ${}^{+}$ Jurkat cells. The expression of a putative RNase5 (angiogenin) receptor [27] has not been described for the Karpas299 cell line, so far.

The CD30-specific scFv-Fc antibody SH313-B5 inhibited the growth of CD30 ${}^{+}$ Karpas299 lymphoma cells with an IC ${}_{50}$ of 100 nM [78]. The immunoRNases 3 and 5 mediated a stronger inhibition of growth of CD30 ${}^{+}$ lymphoma cells than the parental antibody SH313-B5. The immunoRNase1 did not enhance the growth inhibition, which is in contrast to previous data with two other anti-CD30 antibodies [37], but confirms data of IgG-based immunoRNases1 targeting different tumor-associated antigens, which were analyzed for inhibition of tumor cell growth [57]. The immunoRNase3 had the strongest inhibitory effect on the CD30 ${}^{+}$ lymphoma cells with an IC ${}_{50}$ value of $\sim$ 1 nM, which was 100-times enhanced in comparison to the antibody SH313-B5 [78]. Newton et al. [41] measured IC ${}_{50}$ values of 0.2–1 nM for their scFv-RNase2 fusion protein directed towards the transferrin receptor, while free RNase2 mediated no inhibitory effect. Immunotoxins employing CD30-specific antibodies and a mutant of Pseudomonas Exotoxin A achieved IC ${}_{50}$ values between 0.3 ng/mL and 100 ng/mL depending on the target cell line [39]. ImmunoRNases employing a CD22-specific diabody in combination with two Onconase-variants achieved IC ${}_{50}$ values between 3 nM and 20 nM [33].

Since RNase3 did not mediate the highest RNase activity and was at least partially sensitive to the RI, other mechanisms of inhibitory action have to be considered. For example, immunoRNase3 was strongly glycosylated [64], which may enhance cell interaction and internalization. RNase3 is also described to destabilize cell membranes by formation of pores, which can lead to high cytotoxicity [51, 75].

This study further provides the first comprehensive comparison for all catalytically active human RNase A members produced in the same way and using the same substrates and assays. Moreover, the catalytic activities of immunoRNases 1 to 8 and free RNase 1, 4, 5 and 7 were tested in the presence of RI. RI mediated inhibition was quite different among the different members of the gene family. Interestingly, immunoRNases 3, 4, 6 and 7 were only partially inhibited at equimolar RI concentrations, whereas immunoRNases 1 and 2 were not or only slightly inhibited even in 5- to 20-fold molar excess of RI, which draws new light onto the RI:RNase interaction and its biological function. ImmunoRNases 1 to 8 mediated inhibitory effects on CD30 ${}^{+}$ lymphoma cells, which did not directly correlate to the differences in respective RNase activity or the RI inhibition characteristics. This strongly suggests that the processes underlying the cytotoxic activity are dependent on other or at least additional mechanisms. Therefore, in contrast to most of the previous studies usually relying on one antibody moiety and one RNase, future designs of ImmunoRNases would certainly benefit from better clarification of the exact mechanism(s) of action. In addition to the easily accessible biochemical parameters of antibody affinity, RNase activity and RI evasion profile, other and possibly rather complex parameters, like antigen epitope, tissue type and status or internalization pathway usage need to be understood in much more detail before a more general hypothesis for immunoRNase design can be obtained.

4. Material and methods

4.1 Cell lines and cultivation

HEK293-6E cells (NRC, BRI, Montreal, Canada) were cultivated at 37 ${}^{\circ}$ C, 7% CO ${}_{2}$ atmosphere, 95% humidity and 110 rpm in the chemically defined medium Freestyle F17 (Life Technologies, Invitrogen, Carlsberg, CA) supplemented with 25 $\mu$ g/mL G418, 7.5 mM (w/v) L-glutamine and 0.1% (v/v) Pluronic F-68 (Applichem, Darmstadt, Germany) in shaking flasks.

The CD30 ${}^{+}$ ALCL cell line Karpas299 (no. ACC 31, DSMZ, Braunschweig, Germany) and the CD30 ${}^{-}$ acute T cell leukemia cell line Jurkat Clone E6-1 (no. TIB-152; ATCC) were cultivated at 37 ${}^{\circ}$ C, 5% CO ${}_{2}$ atmosphere and 95% humidity in RPMI-1640 (PAA) supplemented with 2 mM (w/v) L-glutamine, 8% (v/v) FCS, and 1% (v/v) penicillin/streptomycin.

4.2 Isolation of RNase genes

The cDNA encoding human RNases 1, 4, 5 and 6 were kindly provided by Zoltán Konthur (Max Planck Institute of Molecular Genetics, Berlin) from the ORFeome library (http://horfdb.dfci.harvard.edu/). The corresponding clones were RNase1 (BC005324, resource 11043), RNase4 (BC015520, resource 11027), RNase5 (angiogenin, BC054880, resource 31022) and RNase6 (BC020848, resource 31018). They were tran-sformed into E. coli XL1-Blue MRF’ (Agilent Technologies, Böblingen, Germany) and selected with 50 $\mu$ g/mL spectinomycin. The cDNA sequences for the human RNases 2, 3, 7 and 8 were amplified from human total cDNA from human peripheral blood leukocytes (Clontech, Saint-Germain-en-Laye, France) by polymerase chain reaction (PCR) using following oligonucleotides: RNase2 (#720 and #721), RNase3 (#722 and #723), RNase7 (#724 and #725) and RNase8 (#824 and #825). The RNases 2, 3 and 7 were cloned by TA cloning into pCR2.1 (Invitrogen) and RNase8 into vector pGEM ${}^{\@setsize{\scriptsize}{8pt}{\viipt}{\@viipt}{\textregistered}}$ -T Easy (Promega, Mannheim, Germany). All oligonucleotide sequences are summarized in Table 1. Correct inserts were verified by DNA sequencing.

4.3 Cloning, production and purification of immunoRNases

The gene fragments of the CD30-specific single chain (sc)Fv antibody SH313-B5 [78] and the DNA encoding the human IgG1-Fc moiety were fused resulting in a sequence encoding a bivalent scFv-Fc antibody. Gene fragments encoding a seven amino acid spacer (GSGGGTS) and the cDNA encoding one the mature human RNases 1 to 8 were added to result in a fusion to the carboxy terminus of the human scFv-Fc antibody in mammalian expression vector pCMV5.2-hIgG1-Fc-XP [45]. Oligonucleotides used for amplification of linker and RNases are summarized in Table 1: RNase1 (#968 and #827), RNase2 (#1118 and #849), RNase3 (#1119 and #851), RNase4 (#1120 and #853), RNase5 (#1121 and #855), RNase6 (#1122 and #857), RNase7 (#1123 and #859) and RNase8 (#1124 and #861). The resulting gene constructs will be referred to as immunoRNases 1 to 8 in this article. In addition to CD30-specific immunoRNases 1 to 8, control constructs were generated using the human scFv-Fc antibody MS112-IIB1 [60] that does not bind to human cells.

The expression was performed in HEK293-6E cells in 500 mL shaking flasks and 100 mL F17 medium. Cells were transfected at cell densities between 1.3 to 2.0 $\times$ 10 ${}^{6}$ cells/mL. Briefly, the DNA of expression vectors was diluted and mixed with polyethylenimine (PEI; Polysciences, Warrington, PA, USA), incubated for 30 min at room temperature and then added to the cells. After 48 h a feeding of 50% (v/v) fresh F17 medium supplemented with 0.75 mL peptone TN1 was done. The immunoRNases were secreted into the culture medium, which was harvested after five days. Protein A affinity chromatography was performed using the Profinia ${}^{\rm TM}$ system, Bio-Scale ${}^{\rm TM}$ Mini UNOsphere Supra and Bio-Scale ${}^{\rm TM}$ Mini Bio-Gel P-6 desalting cartridges (BioRad). Binding-, washing- and desalting-buffer was 137 mM NaCl, 2.7 mM KCl, 4.3 mM Na ${}_{2}$ HPO ${}_{4}$ and 8.1 mM KH ${}_{2}$ HPO ${}_{4}$ . Elution buffer was 100 mM sodium-citrate supplemented with 200 mM NaCl.

The elution fraction was analyzed for purity, integrity and degradation using sodium dodecylsulfate (SDS) polyacrylamide gelelectrophoresis (-PAGE) and Coomassie brilliant blue staining. Concentrations were determined using NanoDropND-1000. All fractions were stored at 4 ${}^{\circ}$ C.

4.4 Cloning, production and purification of free human RNases

The gene fragments encoding the mature form of human RNases 1 to 8, i.e. without leader sequence, were amplified with oligonucleotides according to Table 1: RNase1 (#826 and #827), RNase2 (#848 and #849), RNase3 (#850 and #851), RNase4 (#852 and #853), RNase5 (#854 and #855), RNase6 (#856 and #857), RNase7 (#858 and #859) and RNase8 (#860 and #861) and sub-cloned into the expression vector pYD5 [76] (National Research Council, Biotechnological Research Institute, Montreal, Canada). The resulting gene constructs encoded fusion proteins consisting of human IgG1-Fc part, TEV protease cleavage site (TEVcs) and one of the RNases 1 to 8.

The expression, purification and analysis of elution fractions of the Fc-TEV ${}_{cs}$ -RNase fusion proteins were performed according to the immunoRNases (see above).

TEV protease was kindly provided by Konrad Büssow (Helmholtz Centre for Infection Research, Braunschweig, Germany). The cleavage of Fc-TEV ${}_{cs}$ -RNases was performed with 50 $\mu$ g TEV protease per mg fusion protein for 16 h at 30 ${}^{\circ}$ C. Results were analyzed by SDS-PAGE and coomassie brilliant blue staining. RNase fragments were purified by preparative SEC using Äkta purifier system and the column HiLoad 16–60 Superdex 75 (GE Healthcare) with a flow rate of 0.5 mL/min using PBS as running buffer. The fractions according to the expected molecular mass were collected as 1 mL fractions and analyzed by SDS-PAGE and coomassie brilliant blue staining. RNase fractions with similar purity and concentration were pooled and stored at 4 ${}^{\circ}$ C.

4.5 SDS-PAGE

Protein samples were prepared for 10 min at 95 ${}^{\circ}$ C with Laemmli-SDS sample buffer containing $\beta$ -merc-aptoethanol. Samples were loaded to 15% or 18% SDS polyacrylamide gels. Gels were run with 25 mA and 300 V for 45 min before staining with Coomassie brilliant blue.

4.6 Western blot

Protein samples were prepared for 10 min at 95 ${}^{\circ}$ C using Laemmli-SDS sample buffer containing $\beta$ -merc-aptoethanol. Samples were loaded to 15% or 18% SDS polyacrylamide gels. Gels were run with 25 mA and 300 V for 45 min. Blotting was performed for 35 min at 20 V and 770 mA. The membrane was blocked over night at 4 ${}^{\circ}$ C in PBS containing 2% (w/v) milk powder and 0.1% (v/v) Tween 20 (Sigma, St. Louis, USA) and washed twice with PBS containing 0.1% (v/v) tween 20 for 5 min. Human RNase1 was detected by a rabbit anti-RNase1 serum, which was obtained after immunization with the peptide corresponding to the amino acids 87 to 120 of human RNase1 (Pineda Ab Service, Berlin, Germany) [37]. Detection was done with a goat anti-rabbit IgG antibody alkaline phosphatase (AP) conjugate (Sigma A3687, 1:20,000).

After washing the membrane twice with PBS containing 0.1% (v/v) Tween 20 and once with substrate buffer (100 mM Tris-HCl, 0.5 mM MgCl ${}_{2}$ , pH 9.5) for 5 min, visualization was done by incubation with 1% (v/v) 5-brom-4-chlor-3-indolylphosphat (BCIP, Applichem, Darmstadt, Germany) and 1% (v/v) nitro-blue tetrazolium (NBT, Applichem).

4.7 RNase activity assay

The RNase activity assay was modified from Kelemen et al. [31]. The two different substrates contained 6-carboxyfluorescein (6’-FAM) as fluorophor and Black-Hole-Quencher1 (6’-BHQ1) as quencher. They were linked to the 5’ and 3’ termini of a single stranded chimeric DNA/RNA oligonucleotide sequence, which consisted of one desoxy-adenine (dA), a single uridine (U) or cytidine (C) representing the single scissile bond and two downstream dA. The 6’-FAM-dA-rU-(dA) ${}_{2}$ -6’-BHQ1 (rU) or 6’-FAM-dA-rC-(dA) ${}_{2}$ -6’-BHQ1 (rC) substrates were synthesized by Biomers (Ulm, Germany).

The free RNases or immunoRNases were diluted in 150 $\mu$ L 100 mM 2-(N-morpholino)ethanesulfonic acid (MES) buffer supplemented with 100 mM NaCl in black 96 well microtitre plates. The concentrations were chosen in respect to give optimal kinetic curves. The measurement was started by adding 5 nM substrate in 50 $\mu$ L reaction buffer. Kinetic measurements were done with a Tecan Ultra (Tecan, Crailsheim, Germany) with a gain of 70, a z-position of 7,800 $\mu$ m, an integration time of 40 $\mu$ s and 3 numbers of flashes. The excitation was done at 485 nm and the fluorescence was detected at 535 nm. Measurement was done in 30 cycles à 30 s and additional 60 cycles à 120 s and was finished, when the fluorescence value was constant. The enzyme efficiency ( $k_{cat}$ / $K_{M}$ ) was determined by following equitation: $k_{cat}$ / $K_{M}=$ $v_{0}$ /( $\Delta$ F $\times$ [RNase]) with the initial velocity of the reaction ( $v_{0}$ in $s^{-1}$ ) determined as starting slope of $F(t)$ , the difference between initial and final fluorescence signal ( $\Delta F=$ $F_{max}$ – $F_{0}$ ) and the RNase concentration ([RNase] in M.

4.8 Inhibition of RNase activity by human RNase inhibitor (RI)

Serial dilutions of human RI (RNasin, Promega) were added to the free recombinant RNases or immunoRNases in 150 $\mu$ L 100 mM MES buffer supplemented with 100 mM NaCl to obtain an up to 20-fold molar excess of RI. The mixture was incubated for 15 min at room temperature before the measurement was started by adding 5 nM rU substrate in 50 $\mu$ L reaction buffer. The RNase activity was measured as described above. The formula of the enzyme efficiency was modified to calculate the concentration of active RNase compared to the negative control, which was incubated without RI (Fig. 3: no RI $=$ 100%).

4.9 Size exclusion chromatography (SEC)

The molecular weights of the immunoRNases were determined via analytical SEC with Äkta purifier system (GE Healthcare). The column Superdex200 10/300 GL (GE Healthcare) was used with a flow rate of 0.3 mL/min. PBS was used as running buffer and 100 $\mu$ g of the immunoRNases were analyzed. Absorption at 280 nm was detected.

4.10 Binding studies using flow cytometry

A total of 1 $\times$ 10 ${}^{5}$ CD30 ${}^{+}$ Karpas299 or CD30 ${}^{-}$ Jurkat cells were stained with 100 $\mu$ L 1 $\mu$ g/mL immunoRNase 1 to 8 or the parental scFv-Fc antibody SH313-B5 for 1 h on ice in FACS buffer consisting of PBS supplemented with 2 mM EDTA and 2% (v/v) FCS. After two washing steps with FACS buffer, cells were incubated with 100 $\mu$ L detection antibody goat anti-human IgG (Fc-specific) FITC-conjugated (Dianova 105-095-098; 1:100) for 1 h on ice. Cells were washed two times again and resuspended in 500 $\mu$ L FACS buffer. Measurements were done with a flow cytometer Cytomics FC 500 (Beckman Coulter) with the excitation wavelength of 488 nm and emission wavelength of 525 nm (FL1). A total of 10 ${}^{4}$ events were analyzed per sample. Median fluorescence intensities (MFI) were determined using Flowjo software (Treestar).

4.11 MTT viability assay

A total of 5 $\times$ 10 ${}^{3}$ CD30 ${}^{+}$ Karpas299 or CD30 ${}^{-}$ Jurkat cells per well were seeded in 50 $\mu$ L cultivation medium into flat bottom 96-well plates and cultivated for 18 h at 37 ${}^{\circ}$ C and 5% CO ${}_{2}$ . ImmunoRNases were added to the cells in concentrations ranging from 1 $\times$ 10 ${}^{-7}$ M to 1 $\times$ 10 ${}^{-12}$ M in 50 $\mu$ L medium. All samples were performed as triplicates. Cells were incubated for 72 h. Afterwards, 10 $\mu$ L of a 5 mg/mL 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT, Roth, Karlsruhe, Germany) solution were added to each well and incubated for 4 h along. Reaction was stopped by adding 100 $\mu$ L isopropanol supplemented with 0.04 N HCl. After intensive resuspension, measurement was done using a Multiscan Spectrum Reader (Thermo Electron Corp., Erlangen, Germany) at 570 nm. Dilution series of both cell lines were incubated with MTT substrate and treated the same way to get a calibration for correlation of cell densities and color intensity.

Footnotes

Acknowledgments

We acknowledge Michael Hust and Saskia Helmsing for the provision of the CD30 specific antibody genes. We also thank Konrad Büssow (Helmholtz-Center for Infection Biolology, Braunschweig) for providing the TEV protease and Zoltán Konthur (Drug Response Dx GmbH, Hennigsdorf, and Max Planck Institute of Colloids and Interfaces, Berlin, Germany) for providing RNase 2, 3, 4, 5, and 6 ORF vectors.

Abbreviations:

Appendix

Supplement Fig. 1.

SEC purification of RNase 1 from Fc-TEV ${}_{cs}$ -RNase 1 after cleavage with TEV protease. A) Chromatograms of SEC after cleavage of Fc-TEV ${}_{cs}$ -RNase 1 by TEV protease and the fractions containing the RNase 1 are shown. B) SDS-PAGE analyses of the main fractions of RNase 1 by coomassie staining and C) by Western immunoblot using rabbit anti-RNase1 serum. The two major peak fractions (#50 and 58) correspond to the Fc fragment (#50) and TEV protease (#58) in the chromatogram of SEC.

Supplement Fig. 2.

RNase activity assay with free RNase1 and 5 nM rU substrate in 200 $\mu$ L reaction buffer (black line). Negative control was buffer and substrate without RNase (dotted line). The results are means of triplicates. Measurement was done until a constant fluorescence signal was reached. Starting slope v ${}_{0}$ was determined by linear regression. The maximum increase of fluorescence ( $\Delta$ F) was calculated as difference between initial and final fluorescence signal.

Supplementary data

References

Aleksandrowicz

, Intracutaneous ribonuclease in chronic myelocytic leukemia, Lancet 272 (1958), 420–422.

Atkins

and De Paula

, Physical Chemistry, Oxford University Press 8. Edition, (2006), 810.

Barrabés

Pagès-Pons

Radcliffe

Tabarés

Fort

Royle

Harvey

Moenner

Dwek

Rudd

De Llorens

and Peracaula

, Glycosylation of serum ribonuclease 1 indicates a major endothelial origin and reveals an increase in core fucosylation in pancreatic cancer, Glycobiology 17 (2007), 388–400.

Boix

Vasandani

V.M.

Saxena

S.K.

Ardelt

Ladner

and Youle

R.J.

, Role of the N terminus in RNase A homologues: differences in catalytic activity, ribonuclease inhibitor interaction and cytotoxicity, J Mol Biol 257 (1996), 992–1007.

Böll

Borchmann

and Engert

, Experimentelle Therapie des Hodgkin-Lymphoms, Onkologe 16 (2010), 48–54.

Bosch

Benito

Ribo

Puig

Beaumelle

and Vilanova

, A nuclear localization sequence endows human pancreatic ribonuclease with cytotoxic activity, Biochemistry 43 (2004), 2167–2177.

Braschoss

Hirsch

Dübel

Stein

and Dürkop

, New anti-CD30 human pancreatic ribonuclease-based immunotoxin reveals strong and specific cytotoxicity in vivo , Leuk Lymphoma 48 (2007), 1179–1186.

Cabrita

Gilis

Robertson

Dehouck

Rooman

and Bottomley

, Enhancing the stability and solubility of TEV protease using in silico design, Protein Sci. 16 (2007), 2360–2367.

Carter

Zhang

Dang

T.L.

Hasegawa

Cheng

J.D.

Gianan

O’Neill

J.W.

Wolfson

Siu

Meininger

Kim

Delaney

and Mehlin

, Fusion partners can increase the expression of recombinant interleukins via transient transfection in 2936E cells, Protein Sci 19 (2010), 357–362.

10.

De Lorenzo

Arciello

Cozzolino

Palmer

D.B.

Laccetti

Piccoli

and D’Alessio

, A fully human antitumor immunoRNase selective for ErbB-2-positive carcinomas, Cancer Res 64 (2004), 4870–4874.

11.

De Lorenzo

Di Malta

Cali

Troise

Nitsch

and D’Alessio

, Intracellular route and mechanism of action of ERB-hRNase, a human anti-ErbB2 anticancer immunoagent, FEBS Lett 581 (2007), 296–300.

12.

delCardayre

S.B.

Ribo

Yokel

E.M.

Quirk

D.J.

Rutter

W.J.

and Raines

R.T.

, Engineering ribonuclease A: production, purification and characterization of wild-type enzyme and mutants at Gln11, Protein Eng 8 (1995), 261–273.

13.

Engert

Eichenauer

Dreyling

and Group

E.G.W.

, Hodgkin’s lymphoma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up, Ann Oncol 21 (2010), 168–171.

14.

Falini

Pileri

Pizzolo

Durkop

Flenghi

Stirpe

Martelli

M.F.

and Stein

, CD30 (Ki-1) molecule: a new cytokine receptor of the tumor necrosis factor receptor superfamily as a tool for diagnosis and immunotherapy, Blood 85 (1995), 1–14.

15.

Fang

Jia

K.Z.

Tang

Y.L.

D.Y.

and Hua

Z.C.

, An improved strategy for high-level production of TEV protease in Escherichia coli and its purification and characterization, Protein Expr Purif 51 (2007), 102–109.

16.

Fischer

Nacheva

Mason

D.Y.

Sherrington

P.D.

Hoyle

Hayhoe

F.G.

and Karpas

, A Ki-1 (CD30)-positive human cell line (Karpas 299) established from a high-grade non-Hodgkin’s lymphoma, showing a 2;5 translocation and rearrangement of the T-cell receptor beta-chain gene, Blood 72 (1988), 234–240.

17.

Foyil

K.V.

and Bartlett

N.L.

, Anti-CD30 Antibodies for Hodgkin lymphoma, Curr Hematol Malig Rep 5 (2010), 140–147.

18.

Gerber

H.P.

, Emerging immunotherapies targeting CD30 in Hodgkin’s lymphoma, Biochem Pharmacol 79 (2010), 1544–1552.

19.

Govindan

and Goldenberg

, New antibody conjugates in cancer therapy, ScientificWorldJournal 12 (2010), 2070–2089.

20.

Gupta

Jung

and Brunak

, Prediction of N-glycosylation sites in human proteins, NetNGlyc 1.0 Server; http://www.cbs.dtu.dk/services/NetNGlyc/ In preparation (2004).

21.

Harder

and Schroder

J.M.

, RNase 7, a novel innate immune defense antimicrobial protein of healthy human skin, J Biol Chem 277 (2002), 46779–46784.

22.

Henrichs

Mikhaleva

Conz

Deuerling

Boyd

Zelazny

Bibi

Ban

and Ehrmann

, Target-directed proteolysis at the ribosome, Proc Natl Acad Sci U S A 102 (2005), 4246–4251.

23.

Hetzel

Bachran

Fischer

Fuchs

Barth

and Stocker

, Small cleavable adapters enhance the specific cytotoxicity of a humanized immunotoxin directed against CD64-positive cells, J Immunother 31 (2008), 370–376.

24.

Hofsteenge

Vicentini

and Zelenko

, Ribonuclease 4, an evolutionarily highly conserved member of the superfamily, Cell Mol Life Sci 54 (1998), 804–810.

25.

Holloway

D.E.

Chavali

G.B.

Hares

M.C.

Baker

M.D.

Subbarao

G.V.

Shapiro

and Acharya

K.R.

, Crystallographic studies on structural features that determine the enzymatic specificity and potency of human angiogenin: Thr44, Thr80, and residues 38–41, Biochemistry 43 (2004), 1230–1241.

26.

Horie

and Watanabe

, CD30: expression and function in health and disease, Semin Immunol 10 (1998), 457–470.

27.

Riordan

and Vallee

, A putative angiogenin receptor in angiogenin-responsive human endothelial cells, Proc Natl Acad Sci U S A 94 (1997), 2204–2209.

28.

Hübinger

Müller

Scheffrahn

Schneider

Hildt

Singer

Sigg

Graf

and Bergmann

, CD30-mediated cell cycle arrest associated with induced expression of p21(CIP1/WAF1) in the anaplastic large cell lymphoma cell line Karpas 299, Oncogene 20 (2001), 590–598.

29.

Jäger

Büssow

Wagner

Weber

Hust

Frenzel

and Schirrmann

, High level transient production of recombinant antibodies and antibody fusion proteins in HEK293 cells, BMC Biotechnology 13 (2013).

30.

Jona

and Younes

, Novel treatment strategies for patients with relapsed classical Hodgkin lymphoma, Blood Rev. 24 (2010), 233–238.

31.

Kelemen

B.R.

Klink

T.A.

Behlke

M.A.

Eubanks

S.R.

Leland

P.A.

and Raines

R.T.

, Hypersensitive substrate for ribonucleases, Nucleic Acids Res 27 (1999), 3696–3701.

32.

Klink

T.A.

Vicentini

A.M.

Hofsteenge

and Raines

R.T.

, High-level soluble production and characterization of porcine ribonuclease inhibitor, Protein Expr Purif 22 (2001), 174–179.

33.

Krauss

Arndt

M.A.

B.K.

Newton

D.L.

Seeber

and Rybak

S.M.

, Efficient killing of CD22+ tumor cells by a humanized diabody-RNase fusion protein, Biochem Biophys Res Commun 331 (2005), 595–602.

34.

Küppers

Yahalom

and Josting

, Advances in biology, diagnostics, and treatment of Hodgkin’s disease, Biol Blood Marrow Transplant 12 (2006), 66–76.

35.

Leland

P.A.

Schultz

L.W.

Kim

B.M.

and Raines

R.T.

, Ribonuclease A variants with potent cytotoxic activity, Proc Natl Acad Sci U S A 95 (1998), 10407–10412.

36.

Mathew

and Verma

R.S.

, Humanized immunotoxins: a new generation of immunotoxins for targeted cancer therapy, Cancer Sci 100 (2009), 1359–1365.

37.

Menzel

Schirrmann

Konthur

Jostock

and Dubel

, Human antibody RNase fusion protein targeting CD30+ lymphomas, Blood 111 (2008), 3830–3837.

38.

Mikulski

Costanzi

Vogelzang

McCachren

Taub

Chun

Mittelman

Panella

Puccio

Fine

and Shogen

, Phase II trial of a single weekly intravenous dose of ranpirnase in patients with unresectable malignant mesothelioma, J Clin Oncol. 1;20 (2002), 274–281.

39.

Nagata

Onda

Numata

Santora

Beers

Kreitman

R.J.

and Pastan

, Novel anti-CD30 recombinant immunotoxins containing disulfide-stabilized Fv fragments, Clin Cancer Res 8 (2002), 2345–2355.

40.

Newton

Stockwin

and Rybak

, Anti-CD22 Onconase: preparation and characterization, Methods Mol Biol. 525 (2009), 425–443.

41.

Newton

D.L.

Nicholls

P.J.

Rybak

S.M.

and Youle

R.J.

, Expression and characterization of recombinant human eosinophil-derived neurotoxin and eosinophil-derived neurotoxin-anti-transferrin receptor sFv, J Biol Chem 269 (1994), 26739–26745.

42.

Nichols

N.M.

and Yue

, Ribonucleases, Curr Protoc Mol Biol Chapter 3 (2008).

43.

Ortín

, Immunotherapy of hematological malignancies: what is new? Ann Oncol. 16 (2005), ii53–62.

44.

Palanca-Wessels

and Press

, Improving the efficacy of radioimmunotherapy for non-Hodgkin lymphomas, Cancer Immunol Immunother 116 (2010), 1126–1133.

45.

Pohl

S.C.

Schwarz

Frenzel

and Schirrmann

, A Cassette Vector System for the Rapid Cloning and Production of Bispecific Tetravalent Antibodies, Antibodies 1 (2012), 19–38.

46.

Raines

R.T.

and Ribonuclease

, Chem. Rev. (1998), 1045–1065.

47.

Rehman

Dey

Hassan

Ahmad

and Batra

, Functional role of glutamine 28 and arginine 39 in double stranded RNA cleavage by human pancreatic ribonuclease, PLoS One 6 (2011), e17159.

48.

Ribo

delCardayre

S.B.

Raines

R.T.

de Llorens

and Cuchillo

C.M.

, Production of human pancreatic ribonuclease in Saccharomyces cerevisiae and Escherichia coli, Protein Expr Purif 7 (1996), 253–261.

49.

Rosenberg

and Domachowske

, Eosinophils, ribonucleases and host defense: solving the puzzle, Immunol Res. 20 (1999), 261–274.

50.

Rosenberg

H.F.

, Recombinant Human Eosinophil Cationic Protein, Biological Chemistry 270 (1995), 7876–7881.

51.

Rosenberg

H.F.

, The eosinophil ribonucleases, Cell Mol Life Sci 54 (1998), 795–803.

52.

Rosenberg

H.F.

, Eosinophil-derived Neurotoxin/RNase 2: connecting the past, the present and the future, Curr Pharm Biotechnol. 9(3) (2008), 135–140.

53.

Rybak

Arndt

Schirrmann

Dübel

and Krauss

, Ribonucleases and immunoRNases as anticancer drugs, Curr Pharm Des. 15 (2009), 2665–2675.

54.

Rybak

S.M.

Arndt

M.A.

Schirrmann

Dubel

and Krauss

, Ribonucleases and ImmunoRNases as Anticancer Drugs, Current Pharmaceutical Design 15 (2009), 2665–2675.

55.

Schiffer

Hansen

Hehmann-Titt

Huhn

Fischer

Barth

and Thepen

, Efficacy of an adapted granzyme B-based anti-CD30 cytolytic fusion protein against PI-9-positive classical Hodgkin lymphoma cells in a murine model, Blood Cancer J 3 (2013).

56.

Schiffer

Letzian

Jost

Mladenov

Hristodorov

Huhn

Fischer

Barth

and Thepen

, Granzyme M as a novel effector molecule for human cytolytic fusion proteins: CD64-specific cytotoxicity of Gm-H22(scFv) against leukemic cells, Cancer Lett. 341 (2013), 178–185.

57.

Schirrmann

Frenzel

Linden

Stelte-Ludwig

Willuda

Harrenga

Dübel

Müller-Tiemann

and Trautwein

, Evaluation of human pancreatic RNase as effector molecule in a therapeutic antibody platform, MAbs 6 (2014), 367–380.

58.

Schirrmann

Krauss

Arndt

M.A.

Rybak

S.M.

and Dubel

, Targeted therapeutic RNases (ImmunoRNases), Expert Opin Biol Ther 9 (2009), 79–95.

59.

Schirrmann

Steinwand

Wezler

Ten Haaf

Tur

and Barth

, CD30 as a therapeutic target for lymphoma, BioDrugs 28 (2014), 181–209.

60.

Schütte

Thullier

Pelat

Wezler

Rosenstock

Hinz

Kirsch

Hasenberg

Frank

Schirrmann

Gunzer

Hust

and Dübel

, Identification of a putative Crf splice variant and generation of recombinant antibodies for the specific detection of Aspergillus fumigatus., PLoS One 4 (2009), e6625.

61.

Sharkey

R.M.

and Goldenberg

D.M.

, Targeted therapy of cancer: new prospects for antibodies and immunoconjugates, CA Cancer J Clin 56 (2006), 226–243.

62.

Shih

Wang

and Wang

, Self-cleavage of fusion protein in vivo using TEV protease to yield native protein, Protein Sci. 14 (2005), 936–941.

63.

Sorrentino

, Human extracellular ribonucleases: multiplicity, molecular diversity and catalytic properties of the major RNase types, Cell Mol Life Sci 54 (1998), 785–794.

64.

Sorrentino

, The eight human “canonical” ribonucleases: molecular diversity, catalytic properties, and special biological actions of the enzyme proteins, FEBS Lett 584 (2010), 2194–2200.

65.

Sorrentino

and Libonati

, Human pancreatic-type and nonpancreatic-type ribonucleases: a direct side-by-side comparison of their catalytic properties, Arch Biochem Biophys 312 (1994), 340–348.

66.

Stahnke

Thepen

Stöcker

Rosinke

Jost

Fischer

Tur

and Barth

, Granzyme B-H22(scFv), a human immunotoxin targeting CD64 in acute myeloid leukemia of monocytic subtypes, Mol Cancer Ther. 7 (2008), 2924–2932.

67.

Stein

J.A.

Soldano

A.C.

and Latkowski

J.A.

, Anaplastic large-cell T-cell lymphoma, Dermatol Online J 14 (2008), 15.

68.

Strydom

D.J.

, The angiogenins, Cell Mol Life Sci 54 (1998), 811–824.

69.

Suzuki

Saxena

Boix

Prill

Vasandani

Ladner

Sung

and Youle

, Engineering receptor-mediated cytotoxicity into human ribonucleases by steric blockade of inhibitor interaction, Nat Biotechnol 17 (1999), 265–270.

70.

Tello-Montoliu

Patel

J.V.

and Lip

G.Y.

, Angiogenin: a review of the pathophysiology and potential clinical applications, J Thromb Haemost 4 (2006), 1864–1874.

71.

Terzyan

S.S.

Peracaula

de Llorens

Tsushima

Yamada

Seno

Gomis-Ruth

F.X.

and Coll

, The three-dimensional structure of human RNase 4, unliganded and complexed with d(Up), reveals the basis for its uridine selectivity, J Mol Biol 285 (1999), 205–214.

72.

Tur

Neef

Jäger

Teubner

Stöcker

Melmer

and Barth

, Immunokinases, a novel class of immunotherapeutics for targeted cancer therapy, Curr Pharm Des. 15 (2009), 2693–2699.

73.

Tur

Neef

Jost

Galm

Jäger

Stöcker

Ribbert

Osieka

Klinge

and Barth

, Targeted restoration of down-regulated DAPK2 tumor suppressor activity induces apoptosis in Hodgkin lymphoma cells, J Immunother. 32 (2009), 431–441.

74.

Turcotte

R.F.

Lavis

L.D.

and Raines

R.T.

, Onconase cytotoxicity relies on the distribution of its positive charge, Febs J 276 (2009), 3846–3857.

75.

Venge

and Bystrom

, Eosinophil cationic protein (ECP), Int J Biochem Cell Biol 30 (1998), 433–437.

76.

Wang

Mathias

Stavrou

and Neville

D.J.

, A new yeast display vector permitting free scFv amino termini can augment ligand binding affinities, Protein Eng Des Sel. 18 (2005), 337–343.

77.

Watanabe

Ogawa

Itoh

Koiwa

Kadin

Watanabe

and Okayasu

, Hypomethylation of CD30 CpG islands with aberrant JunB expression drives CD30 induction in Hodgkin lymphoma and anaplastic large cell lymphoma, Laboratory Investigation 88 (2008), 48–57.

78.

Wezler

Hust

Helmsing

Schirrmann

and Dübel

, Human antibodies targeting CD30+ lymphomas, Hum Antibodies 21 (2012), 13–28.

79.

Mikulski

S.M.

Ardelt

Rybak

S.M.

and Youle

R.J.

, A cytotoxic ribonuclease. Study of the mechanism of onconase cytotoxicity, J Biol Chem 268 (1993), 10686–10693.

80.

Zewe

Rybak

S.M.

Dubel

Coy

J.F.

Welschof

Newton

D.L.

and Little

, Cloning and cytotoxicity of a human pancreatic RNase immunofusion, Immunotechnology 3 (1997), 127–136.

81.

Zhang

Dyer

K.D.

and Rosenberg

H.F.

, RNase 8, a novel RNase A superfamily ribonuclease expressed uniquely in placenta, Nucleic Acids Res 30 (2002), 1169–1175.

82.

Zhang

Dyer

K.D.

and Rosenberg

H.F.

, Human RNase 7: a new cationic ribonuclease of the RNase A superfamily, Nucleic Acids Res 31 (2003), 602–607.

Antibody fusion proteins with human ribonucleases 1 to 8

Abstract

Keywords

1. Introduction

Table 1 In this study used oligonucleotides

2.1 Production and purification of immunoRNases

2.2 Production and purification of free human RNases

Table 2 Molecular masses and glycosylation of human RNases

Table 4 Enzyme efficiencies of immunoRNases 1 to 8 and of free RNases 1, 4, 5 and 7

2.5 SEC analysis of immunoRNases

3. Discussion

4. Material and methods

4.1 Cell lines and cultivation

4.2 Isolation of RNase genes

4.3 Cloning, production and purification of immunoRNases

4.4 Cloning, production and purification of free human RNases

4.5 SDS-PAGE

4.6 Western blot

4.7 RNase activity assay

4.8 Inhibition of RNase activity by human RNase inhibitor (RI)

4.9 Size exclusion chromatography (SEC)

4.10 Binding studies using flow cytometry

4.11 MTT viability assay

Footnotes

Acknowledgments

Abbreviations:

Appendix

Supplementary data

References

Table 1
In this study used oligonucleotides

Table 2
Molecular masses and glycosylation of human RNases

Table 4
Enzyme efficiencies of immunoRNases 1 to 8 and of free RNases 1, 4, 5 and 7