Table 1.

CRISPR delivery vehicles and their common features. Relatively difficulty is a subjective measure of how difficult the delivery vehicle is to utilize overall on a four-point scale, where one point is ‘few reagents, facile kit provided’ and four points is ‘requires expert in field with significant experimental experience’.

Delivery vehicle	Composition	Most common cargo	Capacity	Advantages	Limitations	Ease of use	Text refs
Microinjection	Needle	DNA plasmid; mRNA (Cas9 + sgRNA); Protein (RNP)	nM levels of Cas9 and sgRNA	Guaranteed delivery into cell of interest	Time-consuming; difficult; generally in vitro only	****	Yang et al. (2013), Horii et al. (2014), Chuang et al. (2017), Nakagawa et al. (2015), Crispo et al. (2015), Raveux et al. (2017), Sato et al. (2015), Ma et al. (2014), Niu et al. (2014), Wu et al. (2013), Long et al. (2014), Ross (1995)
Electroporation; nucleofection	Electric current	DNA plasmid; mRNA (Cas9 + sgRNA)	nM levels of Cas9 and sgRNA	Delivery to cell population; well-known technique	Generally in vitro only; some cells not amenable	*	Hashimoto & Takemoto (2015), Chen et al. (2016), Qin et al. (2015), Matano et al. (2015), Paquet et al. (2016), Ousterout et al. (2015), Schumann et al. (2015), Wu et al. (2015), Ye et al. (2014), Choi et al. (2014), Wang et al. (2014), Zuckermann et al. (2015), Kim et al. (2014)
Hydrodynamic delivery	High-pressure injection	DNA plasmid; Protein (RNP)	nM levels of Cas9 and sgRNA	Virus-free; low cost; ease	Non-specific; traumatic to tissues	**	Yin et al. (2014), Guan et al. (2016), Xue et al. (2014), Lin et al. (2014), Zhen et al. (2015), Dong et al. (2015)
Adeno-associated virus (AAV)	Non-enveloped, ssDNA	DNA plasmid	<5kb nucleic acid	Minimal immunogenicity	Low capacity	***	Yang et al. (2013), Long et al. (2016), Carroll et al. (2016), Platt et al. (2014), Hung et al. (2016), Swiech et al. (2015), Chew et al. (2016), Truong et al. (2015), Ran et al. (2015), Nelson et al. (2016), Tabebordbar et al. (2016), Esvelt et al. (2013)
Adenovirus	Non-enveloped, dsDNA	DNA plasmid	8kb nucleic acid	High efficiency delivery	Inflammatory response; difficult scaled production	***	Voets et al. (2017), Maddalo et al. (2014), Wang et al. (2015), Ding et al. (2014), Maggio et al. (2016), Li et al. (2015), Cheng et al. (2014)
Lentivirus	Enveloped, RNA	DNA plasmid	∼10kb, up to 18 kb nucleic acid	Persistent gene transfer	Prone to gene rearrangement; transgene silencing	***	Shalem et al. (2014), Wang et al. (2014), Naldini et al. (1996), Kabadi et al. (2014), Heckl et al. (2014); Roehm et al. (2016), Koike-Yusa et al. (2014), Ma et al. (2015), Zhang et al. (2016), Platt et al. (2014)
Lipid nanoparticles/ liposomes/lipoplexes	Natural or synthetic lipids or polymers	mRNA (Cas9 + sgRNA); Protein (RNP)	nM levels of Cas9 and sgRNA	Virus-free; simple manipulation; low cost	Endosomal degradation of cargo; specific cell tropism	**	Yin et al. (2016), Wang et al. (2016), Zuris et al. (2015), Horii et al. (2013), Sakuma et al. (2014), Schwank et al. (2013), Liu et al. (2014), Liang et al. (2015), Kennedy et al. (2014), Miller et al. (2017), Ebina et al. (2013)
Cell-penetrating peptides (CPPs)	Short amino acid sequences	Protein (RNP)	nM levels of Cas9 and sgRNA	Virus-free; can deliver intact RNP	Variable penetrating efficiency	**	Ramakrishna et al. (2014), Axford et al. (2017)
DNA nanoclew	DNA spheroid	Protein (RNP)	nM levels of Cas9 and sgRNA	Virus-free	Modifications for template DNA needed	****	Sun et al. (2015), Sun et al. (2014)
Gold nanoparticles (AuNPs)	Cationic arginine-coated AuNP	Protein (RNP)	nM levels of Cas9 and sgRNA	Inert; membrane-fusion-like delivery	Nonspecific inflammatory response	**	Mout et al. (2017), Lee et al. (2017)
iTOP	Hyperosmlality + transduction compound	Protein (RNP)	nM levels of Cas9 and sgRNA	Virus-free; high-efficiency	Non-specific; no in vivo use yet reported	***	D'Astolfo et al. (2015)
SLO	Bacterial pore-forming toxin	∼100kDa proteins and complexes	Unknown for CRISPR	Reversible pore formation; no impact on cell viability	Not yet proven with CRISPR	***	Sierig et al. (2003), Walev et al. (2001), Brito et al. (2008), Teng et al. (2017)
MENDs	Poly-lysine core, lipid coating, CPP decoration	Nucleic acids	Unknown for CRISPR	Customizable; readily modified for precise delivery	Not yet proven with CRISPR	****	Kogure et al. (2004), Nakamura et al. (2012)
Lipid-coated mesoporous silica NPs	Mesoporous Si coated with lipid	Small molecules and short RNA sequences	Unknown for CRISPR	Inert; easy modification with targeting moieties	Not yet proven with CRISPR	***	Liu et al. (2009), Du et al. (2014), Durfee et al. (2016), Gonzalez Porras et al. (2016), Mackowiak et al. (2013), Su et al. (2017), Wang et al. (2013)
Inorganic NPs	NPs of various compositions (carbon, silica)	Large proteins, nucleic acids	Unknown for CRISPR	Inert; used for similar applications	Not yet proven with CRISPR	**	Bates & Kostarelos (2013), Luo et al. (2014), Luo & Saltzman (2000)