Do you want to build a snowman? Developing and optimizing a qPCR assay to detect ice-nucleating activity

Snowflakes---MA-400x600Over the last few months we have published several blogs about qPCR—from basic pointers on avoiding contamination in these sensitive reactions to a collection of tips for successful qPCR. Today we look in depth at a paper that describes the design and and optimization of a qPCR assay, and in keeping with the season of winter in the Northern hemisphere, it is only fitting that the assay tests for the abundance and identity of ice-nucleating bacteria.

Ice-nucleating bacteria are gram-negative bacteria that occur in the environment and are able to “catalyze” the formation ice crystals at warmer temperatures because of the expression of specific, ice-nucleating proteins on their outer membrane. Ice-nucleating bacteria are found in abundance on crop plants, especially grains, and are estimated to cause one-billion dollars in crop damage from frost in the United States alone.

In addition to their abundance on crop plants, ice-nucleating bacteria are also found on natural vegetation and have been isolated from soil, snow, hail, cloud water, in the air above crops under dry conditions and during rain fall. They have even been isolated from soil, seedlings and snow in remote locations in Antarctica. For the bacteria, ice nucleation may be a method to promote dissemination through rain and snow.

Although ice-nucleating bacteria have been isolated from clouds, ice and rain, little is known about their true contribution to precipitation or other events such as glaciation. Are such bacteria the only source of warm-temperature (above temperatures at which ice crystals form without a catalyst) ice nucleation? Can they trigger precipitation directly? What are the factors that trigger their release from vegetation into the atmosphere? Can we determine their abundance and variety in the environment?

Assay Principle and Considerations

A paper by Hill et al. published in Applied and Environmental Microbiology describes the design of a broad qPCR assay for quantifying bacteria with ice nucleating activity (INA bacteria) on plants and in snow and hail.

Among the challenges for detecting INA bacteria is the fact that in atmospheric samples, INA bacteria can exist as culturable, viable but not culturable (VBNC), moribund and dead cells. Cells falling into any of these categories could potentially drive ice nucleation, particularly cells in the VBNC. For these reasons,any technique that relies on being able to culture these bacteria, would fail to detect cells in the three categories that cannot be cultured.

The assay describe here is designed around the ina gene, which encodes the ice-nucleation protein. A DNA-based assay has limitations: Some dead cells may have degraded their genomic DNA but still possess the INA protein and will go undetected as nucleators, and the presence of the ina gene does not necessarily mean the cell is truly active. However, this assay does overcome many of the limitations of a culture-based survey.

Primer Design

The ina gene has a highly repetitive structure that makes designing a sensitive qPCR assay difficult. The authors chose the region (block 4) of the core of the ina gene that has the least-exact pattern of repetition. This region tends to be more conserved in size among species containing the ina gene, predicting more uniform sizes of PCR products upon amplification, even for previously unidentified alleles.

DNA was isolated from crop and border grasses, snow and hail. First, isolates from all samples were identified based on 16S rRNA gene sequences, amplified from the isolated DNA.
For sequencing block 4 of the ina gene, the authors designed primers based on the published sequences. However, they were unable to use the reverse primer from these initial PCR amplifications for later qPCR because of frequent mispriming. PCR was performed using 1 U GoTaq® Hot Start Polymerase in Colorless GoTaq® Flexi buffer. Reactions contained 1.25mM MgCl2 and 10ng genomic DNA. Among the 20 new isolates obtained from their samples, they were able to identify 18 new alleles of the ina gene from their partial sequences.

For qPCR, primers were designed manually. The authors looked for areas of the INA protein with conserved sequences of amino acids among all known sequences. The sequences that were composed of amino acids with the fewest alternative codons were chosen to minimize primer degeneracy. Because of the repetitive structure of the protein, those primers were evaluated for a unique 3´ end to minimize the possibility of producing multiple products. Several primers were designed and combinations of pairs were tested on isolates. The two best performing primers were used in qPCR assays. Again GoTaq® Hot Start Polymerase and GoTaq® Flexi buffer were used for the qPCR so that Mg2+ concentration could be controlled.

qPCR Optimization

The authors were able to generate qPCR standards using DNA extracted from the INA bacterial species Psuedomonas syringae, which has been completely sequenced, and contains only one copy of the ina gene per cell. Two series of DNA standards extracted from known numbers of cells were produced by extraction using either a normal or high-efficiency method. The normal extraction was used as a standard for qPCR of leaf washings, and the high-efficiency extraction was used as the standard for qPCR of snow samples.

Because the DNA extraction procedure for the snow and hail samples required eliminating the step to remove non-DNA organic compounds or inorganic material and the qPCR required a large volume of the extracted DNA to be added, the authors tested for inhibition of qPCR using spiked replicate reactions.

There is strong secondary structure in the core of the ina gene as a result of a G-C rich sequence every 24 bases. To overcome this secondary structure issue, DMSO was added to the qPCR amplification.

Both of the primer sets tested in the qPCR amplifications were able to amplify most of the alleles of the ina gene, and used together they are predicted to be able to amplify all known alleles, though neither pair was able to amplify all alleles. Both sets amplified undescribed ina gene alleles, and sequencing of qPCR products from  leaf washings of additional sampled crops revealed other alleles that were not found in cultured isolates screened from the same leaf washings (unpublished data), suggesting the strains bearing these alleles were not easily cultured.


The authors of this study set out to create a broad assay to identify INA bacteria from environmental samples that would overcome the limitations of bacterial surveys that require culturing bacteria. In doing so they optimized a PCR assay that was able to detect previously undescribed alleles of the ina gene. This not only provides a tool for better understand the prevalence and biology of INA bacteria, but also provides a nice example of the kinds of considerations that go into optimizing a qPCR assay.

The following two tabs change content below.

Michele Arduengo

Social Media Manager at Promega Corporation
Michele earned her B.A. in biology at Wesleyan College in Macon, GA, and her PhD through the BCDB Program at Emory University in Atlanta, GA. Michele manages the Promega Connections blog. She enjoys leisure reading, writing creative nonfiction and knitting, and the occasional cross-country skiing jaunt.

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.