Tuesday, April 04, 2017

SageHLS: Automated uHMW DNA Preparation

Advances in optical mapping, linked reads, PacBio and nanopore sequencing are enabling generating highly contiguous large genome sequences routinely and inexpensively.  However, this in turn is creating intense demand for efficiently and reliably preparing ultra-high molecular weight (uHMW) DNA.  By this term,  I mean DNA approaching or exceeding a megabase in size.  Methods for preparing HMW and uHMW DNA tend to be very old-school, reaching back at least back to the 1970s, 80s and 90s for approaches used in the early days.  Phenol-chloroform preps with the DNA spooled out onto a glass hook or rod are one popular approach; another is to embed cells in agarose blocks, extract the DNA within the block and then degrade the agarose to retrieve the DNA.  Nuclei preps are yet another approach. Any liquid handling must be performed gently and with wide bore pipettes.  These techniques tend to be tedious and slow affairs, requiring many manual steps.  As an alternative, Sage Sciences has launched an instrument which automates a process with no hazardous chemicals, the SageHLS.

A group from Sage hosted me recently for a tour of their facility north of Boston (it's actually the same space, for better or worse, where the SOLiD instrument was developed) and a detailed chat about the SageHLS and associated protocols. Images shown below are all courtesy of Sage.  Sage has been in the electrophoretic sample prep business since the previous decade, shifting with changes in customer preferences.  For example, their original Pippi Prep instrument was engineered for size selecting typical short read libraries, replacing manual agarose slab gels and the associated tasks of extracting the sized DNA.  After successfully selling a number of units, demand slackened as SPRI bead size selections came online.   New demand for this instrument has come from sizing microRNAs and more recently for nucleosome-sized and smaller fractions of cell-free DNA.

SageHLS is their newest instrument, launched about a month ago.  For a list price of $35K and with a benchtop footprint similar to other small equipment, the SageHLS has a capability of extracting DNA from four samples per workday, with workflows requiring minimal human intervention and typically lasting 5-7 hours.  Two basic classes of workflows are currently offered: bulk DNA extraction with optional size selection and CATCH.  CATCH is a CRISPR-assisted scheme for capturing targeted DNA.

The device can take two cartridges, each of which has two sample sections.  The agarose for SageHLS gels are poured by robots attended by gowned employees; cartridges for some of the older systems are poured manually in a hood.  Each cartridge has two wells of approximately 80uL volume.  One is loaded with cells or spheroplasts (cell walls must be stripped off-line) in an isotonic and isopycnic solution to ensure that cells are not lysed early and that they are uniformly distributed throughout the sample well.  The other well is loaded with a lysis buffer containing a high concentration of SDS.  When the program starts, the SDS is electrophoretically swept into the sample well.  After incubation, electrophoresis is used to move all of the charged components into the gel.  Proteins, RNA, ions and such migrate through the gel, but the uHMW DNA is trapped in the agarose.  

For bulk DNA isolations, a dilute solution of NEB's Fragmentase, an enzyme cocktail, is added to break the DNA into fragments which can be electrophoresed out of the gel and onto a ultrafiltration membrane in one of the elution wells of the cartridge.  Size selection can be applied, though at this time it works best for "shorter" DNA (sub-500kb) than to select for larger DNA.  Alternatively, in the CATCH protocol Cas9 with appropriate guide RNAs is used to liberate only the DNA region of interest.  A longer-term goal is to perform complete library preparations in the device, with Oxford Nanopore's transposase-based Rapid1D method being an obvious target.  The instrument has controllable heater elements to support different incubation temperatures for different protocols. Sage is envisioning standard preps to cost about $50 per sample, with CATCH preps targeting $100 per sample.

Sage reports that some of their collaborators in the optical mapping space have observed 2 megabase DNA molecules extracted by the SageHLS.  Sage is working on improving the quantity of DNA (and therefore concentration) coming from the machine,  Elution is currently into about 80 ul of Tris-TAPS buffer, which has the interesting property that the two ions have opposite electrophoretic mobilities of the same absolute magnitude. An example of a CATCH elution is shown in the second image below; note the bright band at the target size.

With ultralong reads and maps becoming a very hot topic, it will be interesting to see how many groups stick with the well-tested old school methods and which take the plunge on the SageHLS.  Compared to the cost of an optical mapping instrument of a PacBio Sequel, the Sage is a somewhat modest further investment.  Certainly a lab performing many uHMW preps could justify the labor savings from this instrument.  So even though SageHLS's price dwarfs that of a MinION, using skilled lab personnel more efficiently counts for a lot.

Datasets generated with SageHLS aren't yet publicly available.  A key question will be the uniformity of coverage; does the Fragmentase fall prone to hypersensitive spots (with subsequent drop-out of information) or deserts which don't cleave well?  Does this need to be tuned for different organisms? NEB makes claims for uniformity, but these were looking at generating short insert libraries for short read sequencing; it is conceivable that patterns not visible at that scale could emerge with long reads.

With the ability to process 4 samples per day, SageHLS should be able to keep up with optical mapping instruments.  Oxford's new GridION X5 can run 5 samples, but with runs of two days (probably soon extending into a third), SageHLS would still be able to keep up. This is particularly true for human-sized genomes, as it takes multiple MinION flowcells to generate enough coverage, though for sequencing most fungi and smaller invertebrate genomes (such as Drosophila or Caenorhabiditis and many others), a one genome per flowcell pace could be reasonably expected.  However, now the first tweet has emerged claiming PromethION data at a customer site, the spectre of high throughput sequencing on a previously unimaginable scale looms and with it small corrals of Sage instruments might be needed.

This instrument also underscores that nucleic acid extraction and library prep remain a hot topic, with all sorts of permutations of different technologies being tried out.  Don't be surprised to see more dispatches in this space exploring further innovations in getting from biosample to the flowcell.

No comments: