Current research (November 2023 – June 2026)

A new project has recently commenced (November 2023) that aims to facilitate the mass-rearing, release, monitoring and evaluation of the impacts of the rust fungus (Puccinia rapipes) for the biological control of African boxthorn (Lycium ferocissimum) across NSW. For more information on our annual project updates, please scroll further down to the “Annual Update” heading. 

This new project is proudly supported by the NSW Government through its Environmental Trust.

Project Overview

Output 1, monitoring and evaluation: The project will enable the establishment of a network of biocontrol agent release sites throughout NSW, in which the performance of the fungus and its effects on the target weed will be monitored regularly over time.

Output 2, community engagement: The project will also deliver the fungus and release instructions to key weed stakeholders across NSW, targeting areas where African boxthorn exerts the greatest threat to biodiversity values. Participants may include, but not limited to, Aboriginal and Torres Strait Islander peoples, biosecurity practitioners (e.g., council weed control officers), volunteer community groups (e.g., Landcare, Bushcare), Rangers (e.g., NPWS), private landholders and other key stakeholders with interests in African boxthorn biocontrol.

This project will include the following main activities:

• Maintain and mass-rear a healthy, viable culture of the biotrophic rust fungus:
  • To be achieved under laboratory conditions on healthy, pest-free African boxthorn plants (based at the CSIRO Black Mountain Laboratories in Canberra, ACT).
• Deliver the fungus to registered stakeholders for release into the environment to help control African boxthorn infestations:
  • It is proposed that the fungus will be delivered to up to 90 key stakeholders altogether over three years (resulting in release of the fungus at up to 260 sites). All stakeholders with interests in African boxthorn control are eligible to register interested in participating in the project, although there is no guarantee that all parties will receive the fungus if demand for participation is greater than our ability to culture the fungus in the laboratory.
• Establish up to 10 fixed African boxthorn monitoring sites throughout NSW:
  • The CSIRO field ecologists will collect baseline weed population data at sites located across a broad climate gradient, with priority given to infestations located in grassy woodland communities in conservation reserves that are most under threat by African boxthorn invasion (especially along creeklines and other water sources).
• Release the fungus at the fixed monitoring sites under optimal climate conditions:
  • Releases will preferentially be made during spring 2023, but summer releases will also be considered during periods of high rainfall (or high humidity) under moderate daytime temperatures (< 30°C).
• Monitor the performance of the fungus and its effects on the host African boxthorn plants at the fixed monitoring sites over time:
  • The fungus will be released using optimised methods developed by CSIRO researchers. Post-release monitoring will be undertaken once every six months (or at least once per year) to evaluate establishment and spread of the fungus, disease incidence and severity on the host plants, and changes in the cover and condition of African boxthorn foliage in response to fungal infection.

Background information on the fungal biocontrol agent, Puccinia rapipes

The biocontrol agent is a biotrophic rust fungus, Puccinia rapipes, that infects the leaves of African boxthorn. It was originally isolated from diseased boxthorn plants in South Africa. Through extensive host-specificity studies undertaken by the CSIRO, the fungus was shown to be highly specific to African boxthorn and other introduced, closely related Lycium species, and poses no danger to native Lycium species and other Australian species. In 2021, the fungus was approved for release into the Australian environment as a biocontrol agent to assist with the control of African boxthorn.

The full data set and comprehensive risk analysis are provided at this link: https://www.agriculture.gov.au/biosecurity-trade/policy/risk-analysis/biological-control-agents/risk-analyses/completed-risk-analyses/puccinia-rapipes

The rust fungus infects young leaves of African boxthorn, causing yellowing of the leaves followed by the development of pustules. The pustules produce fungal spores which are dispersed by wind. The spores land on the leaves of nearby African boxthorn plants and, under humid conditions, will germinate and infect new leaves. Infected leaves will die back over time. This may result in extensive defoliation of an individual plant if the fungus establishes widely and causes severe disease. Infection by the rust fungus can also disrupt the photosynthetic capacity of the plant, reducing overall plant growth and reproductive output.

Releasing the biocontrol agent onto African boxthorn

Between November 2023 and June 2026, the CSIRO will provide participants in NSW with a biocontrol agent release kit that contains one or more vials of rust fungal spores. To apply the fungus to African boxthorn plants, participants will mix the dried spores with TWEEN 80 and water in a plastic spray bottle provided. This solution of spores and water will then be misted onto the leaves of the boxthorn plants in the late afternoon, and the treated stem covered with a plastic bag overnight. The fungal spores will germinate in this humid microenvironment and infect the leaves of healthy plants. The plastic bag will then be removed the following morning to prevent the fungus from being killed be elevated temperatures inside the bag when exposed to daylight.

Typical symptoms of infected boxthorn leaves and fruit in the field.

Eligibility to receive the biocontrol agent

The fungus is only available for release in NSW at this stage, as part of this current project supported by the NSW Environmental Trust, up until June 2026. Engagement will be targeted at community and public sector organisations, such as (but not limited to) Landcare, Local Land Services, NSW, Bushcare, local governments, other volunteer groups etc. Private landholders are also eligible to express an interest to participate in the program. The fungus is not commercialisable and cannot be released as part of for-profit ventures, including retail or as part of consultancies.

Please email Caroline Delaisse at boxthornbiocontrol@csiro.au to register your interest to participate in the biocontrol program. The CSIRO will endeavour to respond within one week of receiving the email. As part of the registration, the CSIRO will provide a pre-release assessment document for participants to complete, which requires information about the proposed location of the boxthorn infestation. The pre-release assessment needs to be undertaken prior to receiving the agent.

The biocontrol agent will be provided to registered participants at regular intervals between November 2023 and June 2026. The specific timing of delivery will be determined by prevailing climate conditions, quantity of available material for release and level of demand from registered participants.

The release kit will be sent via post. Participants should release the agent as soon as possible upon receiving their package. However, the spores of the biocontrol agent can be stored in the freezer for one month before preparing the biocontrol agent for release onto African boxthorn plants in the field if conditions at the time of receipt are unsuitable.

FAQs

Is it safe to release the biocontrol agent into the Australian environment?

In 2017, CSIRO began rigorous evaluation of the risks that the fungus Puccinia rapipes could pose to non-target plants in Australia. Research focused on species within the family Solanaceae. This extensive host-specificity testing was performed in a quarantine facility and involved exposing African boxthorn and non-target plant species to the fungus under optimal conditions for infection. It was found that the fungus is highly specific to African boxthorn (Lycium ferocissimum) and Goji berries (L. barbarum, L. chinense and L. ruthenicum; of which the rust can be easily treated with fungicides already used by goji growers). Based on these research results and following a comprehensive risk assessment process and public consultation, the Federal Department of Agriculture, Water, and the Environment (DAWE) approved the release of the biocontrol agent into the Australian environment. The information package that supported the application to release the agent in Australia, which includes all results, can be found here: https://www.awe.gov.au/biosecurity-trade/policy/risk-analysis/biological-control-agents/risk-analyses/completed-risk-analyses/puccinia-rapipes

What quantity of the biocontrol agent will I receive for release into the environment?

Registered participants will receive via post a “biocontrol agent release kit”. Each biocontrol agent release kit will contain a plastic vial of dried spores collected from pustules produced by the rust fungus. This vial of spores can be used to make a release at one site on a small number of African boxthorn plants. As those plants become infected, the fungus develops pustules which produce spores. Those spores will in turn naturally spread to nearby plants via wind or water and continue the spread of the fungus.

CSIRO staff will work with you to determine the total number of release kits provided, depending on your capacity/interest to release the fungus in your local region. To increase the chances of successful establishment of the fungus, we encourage participants to release the fungus at two points with two separate vials of spores, each separated by at least 200 m. Participants are welcome to request more release kits, but supply is dependent on stock availability. 

What effect will the biocontrol agent have on African boxthorn populations? How quickly does the fungus kill African boxthorn?

In the Australian environment, the fungus is not expected to kill African boxthorn. Provided that the biocontrol agent establishes widely and causes severe disease symptoms on African boxthorn, it is expected to reduce the reproductive output and vegetation growth of the weed in the long term. This will in turn reduce its invasion potential in various ecosystems, but will not eradicate it altogether.

How quickly will the fungus spread?

The rate of spread of the fungus will be ascertained by long term monitoring. It is expected that the fungus will spread from one plant to the next very slowly, but the rate of spread will accelerate once the overall abundance of the fungus builds up in the local African boxthorn population. Based on our knowledge of other successful biocontrol agents that have been released previously in Australia, broadscale spread of the fungus would be expected to take several years and will not occur within the first season of release. As such, ‘success’ in the short term for this research project is to first establish the presence of the fungus in the Australian environment.

Can use of the biocontrol agent replace herbicide application or other control methods?

Biocontrol may provide a sustainable, landscape-scale approach for African boxthorn management with no chance of off-target damage to crops or native vegetation. Given that the fungus will not kill African boxthorn altogether, it will complement but not replace the need for other control methods. However, widespread establishment and spread of the fungus may gradually reduce the quantity of chemical herbicide required to suppress weed populations.

What happens if I cannot detect the fungus after release? Has it failed?

It is important to note that the fungus will only infect African boxthorn at high severity under optimal conditions for growth and spread – that is, when warm and moist and the host African boxthorn plants are healthy and vigorous at early stages of growth. As such, we expect that the fungus will not establish in all instances. The fungal spores are delicate and require specific microclimate requirements for germination and infection. The fungus will only become widely established in the Australian environment after many years of sustained releases by various participants.

As such, participants are encouraged to release the fungus on multiple occasions where the initial releases may have failed.

Additional documents

Registration document (for interested participants to fill in prior to receiving the biocontrol agent)

Release instructions on to release the biocontrol agent

Annual Update: November 2023 – December 2024

The project was officially launched in November 2024 with an announcement by the NSW Minister for the Environment, Penny Sharpe.

In an online article entitled “Rust is a must for African Boxthorn battle”, the CSIRO explains the aims and aspirations of the project, and how the community, public sector weed managers and conservation groups can get involved with releasing the rust fungus:

• Engage all stakeholders with interests in the control of African boxthorn in the biocontrol release program, including (but not limited to) private landholders, public sector land managers (e.g., LGA weed biosecurity officers), Landcare groups, etc. Engagement will occur through in-field demonstrations and workshops, online through annual email updates using the project account (boxthornbiocontrol@csiro.au) and printed information sheets.
• Distribute biocontrol agent release kits, along with release and monitoring instructions, to registered participants throughout NSW.
• Experimentally release the fungus at CSIRO-managed fixed monitoring plots, and evaluate the establishment, spread and impacts of the fungus on the host weed populations.

 

Workshops

Altogether, the CSIRO delivered 12 workshops during this first year, represented predominantly by Local Land Services, local governments and Landcare (see table below). The workshops were delivered to approx. 200 participants and consisted of a combination of presentations/seminars, in-field demonstrations of biocontrol agent release methods, and training in monitoring and evaluation techniques. Attendees were provided with printed copies of the presentation slide deck, program registration forms and the biocontrol agent release, monitoring and evaluation methods.

Information on the 10 workshops delivered for the African boxthorn biocontrol project.

Date	Location	Activity	Partner organisation	No. attendees
9th October 2023	Dubbo	Field demonstration of fungal release methods; training in monitoring and evaluation method	Dubbo Regional Council	2
11th October 2023	Forbes	Field demonstration of fungal release methods; training in monitoring and evaluation method	Forbes Shire Council	5
27th November 2023	Hay	Field demonstration of fungal release methods	Hay Shire Council, Riverina Local Land Services	2
22nd February 2024	Online	Information session	Hunter Local Land Services, Martindale Landcare	5
6th March 2024	Warren	Face-to-face seminar and field demonstration of fungal release methods	Central West Local Land Services, Castlereagh Macquarie County Council	~50
7th March 2024	Coonamble	Face-to-face seminar and field demonstration of fungal release methods	Central West Local Land Services, Castlereagh Macquarie County Council	~50
13th May 2024	Nyngan	Face-to-face seminar and field demonstration of fungal release methods	Bogan Shire Council, Hilltops Council, Macquarie Valley and Lachlan Valley Weeds Committees	~10
15th May 2024	Forbes	Training in monitoring and evaluation methods	Forbes Shire Council	5
21st May 2024	Deniliquin	Face-to-face seminar and field demonstration of fungal release methods	Murray Local Land Services, Edward River Council	~30
22nd May 2024	Hay	Training in monitoring and evaluation methods	Hay Shire Council, Riverina Local Land Services	2
9th October 2024	Upper Hunter	Face-to-face seminar and field demonstration of fungal release methods	Hosted by Martindale Creek Catchment Landcare, supported by Hunter Local Land Services and Upper Hunter Weeds Authority	~10
24th October 2024	Condobolin	Face-to-face seminar and field demonstration of fungal release methods	Central West Local Land Services	~30

 

 

Biocontrol agent delivery to community stakeholders

Since the program was launched, 170 community members have expressed interest in the program, of which 50 subsequently registered to receive and release the biocontrol agent into the environment. This included 35 private citizens and representatives from 11 regional councils, Landcare and Local Land Services. Altogether, the 50 participants have registered 98 sites at which the fungus will be/has been released at (see figure below).

Between December 2023 and August 2024, the CSIRO dispatched 102 biocontrol agent release kits to 33 of the registered participants, who to date have released the biocontrol agent at 41 sites using the approved methods.

The CSIRO will continue to register interested participants for ongoing releases of the fungus in autumn 2025. There remain significant gaps in the distribution of the fungus, especially in the Sydney Basin, Hunter, New England and Murray River. We invite participants from these regions to assist by releasing the fungus in autumn 2025.

 

Monitoring plot establishment

In October and November 2023, the CSIRO set up 18 permanent monitoring sites at 9 localities and released the fungus at a subset of 16 of these sites. Sites were set up in dense infestations of African boxthorn in remnant native woodlands with an open Eucalyptus/Callitris canopy and understorey of mixed native/exotic perennial pastures (example photographs in figure below). We avoided heavily grazed areas that were disturbed by farm infrastructure or weed control activities. Sites were spread across a broad region of NSW, from the Central West, Southern Ranges, Riverina and Murray River.

At each site, up to 20 replicate African boxthorn plants were selected with each plant separated by ~2m. We selected plants that were reproductively mature with vigorous, lush and green foliage with no signs of physical damage or physiological stress.

Half of the plants were allocated to the fungal-inoculation treatment whilst an equal number of plants of similar size (height, diameter, foliage cover) were allocated to the non-inoculated control. For each plant, baseline measurements were taken on plant size (height, diameter) and foliage condition (% cover and the number of leaves, flowers and fruit on four 10-cm stem lengths selected at random per branch per plant).

Immediately prior to sunset, a single large branch on each of the fungal-inoculation treatment plants was sprayed liberally with a fungal spore suspension (inoculum was made up of 0.1ml spores in 250 ml 0.1% TWEEN 80; each branch was sprayed with 25 ml of inoculum), then covered in a clear plastic bag to maintain a humid environment overnight (see methods developed previously by the CSIRO: https://research.csiro.au/african-boxthorn/achievements-rrnd4p-rnd-2/). The bags were removed early the next morning at sunrise. The experimental control plants were sprayed with the same volume of solution (water mixed with Tween minus the fungal spores) and covered in plastic bags to control for the effects of bagging on foliage development.

Microclimate data (temperature and relative humidity) were recorded using 36 iButton dataloggers deployed at 18 release sites. Of the 36 deployed iButton dataloggers, data were retrieved from 14 iButtons from 11 release sites for the 2022/23 monitoring period. Microclimate variables were monitored hourly during the bagged releases (which were made between October-November 2023) and then for up to six-months (datalogger storage capacity) until the sites were monitored again in Autumn 2024. Of the 14 dataloggers, 11 dataloggers from 9 release sites recorded microclimate data during the bagged releases (note, 3 iButtons were missing data). Data revealed that the average temperature overnight during the inoculation event ranged between 10-20 °C across the 9 sites, and average relative humidity was > 90 % at 7 of the 9 sites. These microclimate ranges are suitable for germination of the fungal spores and infection of the host plants during 12 hr.

The microclimate data were then evaluated over the entire post-release monitoring period (Summer 2023 to Autumn 2024). In the Booroorban and Deniliquin region, releases were made in late November 2023 and monitored ~6 months later in late May 2024 whereas in Dubbo and Forbes, releases were made in mid-October 2023 and monitored ~7 months later in mid-May 2024. The data revealed that there was considerable variability in both temperature and relative humidity within each site over time, but altogether the average daily humidity was ~ 50 % and average daily temperature ranging from 20-30 °C. At all sites, average maximum daily temperatures over the entire recorded monitoring period were 30 °C or greater and regularly exceeded 40 °C during summer.

 

Biocontrol agent establishment

In May 2024, approximately six months following the initial releases of the fungus, we returned to 12 of the monitoring sites to evaluate establishment of the fungus and its effects on the host plants. Visual symptoms of infection by the fungus (i.e., raised lesions on the leaf surface) were not detected on any of the inoculated plants at any of the release sites.

Microclimate data verified that the releases were made during optimal conditions for spore germination and infection of the host plants, and the host plants were healthy and suitable as hosts. We speculate that the very high average and maximum daily temperatures experienced at all sites within the first week following release, followed by sustained high temperatures and very low rainfall at most sites over summer, inhibited infection development by the fungus and establishment within the resident African boxthorn populations. We also received reports from several partner organisations that the very hot and dry temperatures resulted in broadscale defoliation and dormancy of the African boxthorn plants across each region. The rapid shedding of leaves and then subsequent regeneration of new leaves in autumn 2024 would likely have prevented us from detecting the fungus, even if initial infection was successful. As such, during the follow-up surveys in May 2024, we released the fungus a second time at a subset of six sites (Table 2).

When we returned to the plots in May 2024, we released the fungus a second time at 6 plots, using the same methods as described above. We returned to all plots again in October 2024. We observed spectacular results! The fungus was detected at high abundance and disease severity at 100 % of the plots where it was released a second time in May 2024 (see photos below). This suggests that there are strong seasonal effects on establishment of the fungus, and that autumn is a more optimal time of the year to release it into the field. At these sites, we also detected the fungus on nearby plants that were not originally inoculated, meaning the fungus has begun spreading in the local environment.

Follow-up surveys in 2025 will allow us to determine how quickly the fungus can spread and indeed whether it is able to persist in the local environment over summer. We expect that such persistence is likely, given we found the fungus at very density at 17 % of sites where it was only released once in 2023.

 

Impacts of the fungus on African boxthorn populations

So far, infection by the fungus, including in areas with high disease severity, has not resulted in significant decline in the foliage cover of the host weed. This is expected, as biocontrol agents will only exert significant impacts on the host weed once they have become well established in the environment at high density and spread widely across the landscape. Even for some of Australia’s most successful biocontrol projects, benefits through reduced weed abundance are realised over years to decades. As the rust fungus is not a bioherbicide, it is not expected that infection will result in defoliation of African boxthorn following its release. As such, the primary aim of this project is to widely release the fungus across NSW and get it well-established in the NSW environment, sustained over time. Ongoing research on African boxthorn foliage, reproductive output and populations in response to infection by the fungus will, in time, reveal its impacts on the host plant. For now, the fact that we have been able to induce infection on naturalised populations of Africa boxthorn in the field for at least one year is considered a resounding short-term success