Rag1D600A, a novel catalytically inactive RAG mouse model

The RAG complex (RAG1 and RAG2) can bind to recombination signal sequences of antigen receptor loci gene segments and coordinate V(D)J recombination which is the primary method of generating antigen receptor diversity. Previous biochemistry studies discovered RAG1 D600, D708 and E962 residues as essential for catalytic DNA nicking and hairpin forming activity of the RAG complex. Neutralization of each of the acidic residues does not impair DNA binding to recombination signal sequence containing DNA substrates, but cleavage of the substrates is severely compromised. These three acidic residues are thought to comprise a DDE motif that is responsible for binding to a divalent cation that is necessary for cleavage activity. Although a Rag1-/-; RAG1-D708A transgenic mouse model system has been used to study dynamics of RAG activity, transgenic expression may not precisely mimic expression from the endogenous locus. In order to improve upon this model, we created Rag1D600A mice that lack B and T cells and demonstrate a developmental block at the pro-B and DN stages, respectively. Thus, Rag1D600A mice provide a novel mouse model system for studying the poorly understood noncanonical functions of RAG1.


43
Two RAG1-RAG2 monomers form a homodimer collectively called the RAG complex.

55
The RAG2 core region comprises an N-terminal WD40 domain that folds into a b-propeller [8].

56
To further dissect the mechanisms of how RAG catalyzes DNA cleavage, highly 57 conserved amino acids were individually analyzed in RAG1. Three important acidic residues, 58 D600, D708 and E962 were identified. When each of these were individually mutated to a 59 nonpolar alanine (A), DNA cleavage activity was severely compromised although RAG retained 60 its ability to bind to the RSS of a DNA substrate [9][10][11]. This highlights a crucial role for these 61 three residues in RAG cleavage activity.

71
The early crystal structures of the RAG complex revealed that it takes on a "Y" shape,

72
where each of the two RAG1 proteins bind to each other at the base via the DDBD and RAG2 73 sits on top of the arms that extend up from the base [15]. Advances in structural analysis of 74 RAG revealed that the three RAG1 acidic residues (D600, D708, E962) are located in close 3D 75 proximity to one another [16,17]. The colocalization of the three essential acidic residues (D600, D708, E962) at the active site of RAG1 provides further support that this is a DDE motif. RAG 77 has a strikingly similar structure to two other DDE family transposases: the well-known bacterial 78 transposase, Tn5 and the hAT family transposase, Hermes [16,18].

79
The combination of early RAG1 mutational, DNA sequence and protein structural 80 analyses strongly support the idea that D600, D708 and E962 of RAG1 comprise a DDE motif 81 that is essential for executing RAG mediated DNA cleavage in a similar manner to DDE family

107
To circumvent these issues, we used CRISPR-Cas9 technology to make a point

118
Zygotes were cultured to the blastocyst stage and then transplanted into host mice. After pups 119 were born, tail tissue was collected and founder mice were assessed by PCR. The ssODN 120 repair template was designed to incorporate five different point mutations around the site 121 encoding the D600 residue of RAG1. To prevent the sgRNA from cutting the DNA after repair, 122 two silent mutations upstream of the D600 site were introduced to disrupt the PAM-sequence 123 and sgRNA binding capability. The GAT nucleotides that encode the aspartic acid (D) at residue 124 600 were converted to GCC to encode an alanine (A). In order to be able to easily detect alleles 125 with the point mutation, one additional silent mutation was designed three DNA base pairs 126 further downstream to create a NaeI/NgoMIV restriction site. Founder mice were identified by PCR and sequenced to confirm that all the mutations from the ssODN were incorporated (Fig   128   1A). To detect Rag1 D600A alleles by PCR, primers were designed to create two different sizes 129 following restriction enzyme digest of the PCR product. Failure to digest the PCR product, 130 partial digestion or complete digestion were respectively indicative of wild-type, Rag1 D600A/+ , or 131 Rag1 D600A/D600A mice (Fig 1B).

141
Since V(D)J recombination is essential for driving B and T cell development, it is 142 expected that Rag1 D600A mice mice will have a deficiency in B and T cells. To assess this, thymi 143 and spleens that harbor developing T cells and mature B cells respectively were dissected from 144 wild-type and Rag1 D600A mice. The Rag1 D600A thymi and spleen were decreased in size 145 compared to wild-type littermate controls, implying a severe immune cell deficiency (Fig 2).

161
To determine if B cell development is also blocked at the pre-B cell stage due to a failure 162 in Igk rearrangement, we crossed Rag1 D600A mice with mice containing a pre-rearranged heavy-163 chain transgene (B1.8). Rag1 D600A ; B1.8 mice also failed to make mature and immature B cells, 164 exhibiting a developmental block at the pre-B cell stage (Fig 4). Thus, the B1.8 allele allows B   (Fig 1.). To examine this further, we analyzed T cell development from the thymus by flow

199
Although we attempted to make both Rag1 D600A and Rag1 D708A mice, we failed to recover 200 any founder mice that harbored the Rag1 D708A allele. It is unclear why this is the case, but it is 201 possible that the sgRNA designed for the D708 site had off target binding activity that 202 decreased the efficiency of editing at the site of interest. Since we obtained founders for 203 Rag1 D600A , we did not continue further with the D708A design.

223
Mouse zygotes were injected with the following mix: in vitro transcribed 50ng/uL sgRNAs, 224 100ng/uL Cas9 mRNA (Trilink L-6125) and 50ng/uL ssODN (IDT custom oligos). Target 225 sequences were cloned into a Px461 plasmid using cut and paste cloning with a BbsI restriction 226 enzyme. T7 promoter sequences were added by PCR and the PCR product was used as 227 template to generate sgRNAs by in vitro transcription using the HiScribe T7 quick high yield and oligos used are listed in S1 Table.