Weeds are plants that grow in a location where their presence has become undesirable due to their adverse effects on the ecosystem or human activities (Norris 1992; Kent 1994; Gadermaier et al. 2014). This is mainly due to the fact that when a plant is introduced to a new environment without the influence of their natural enemies they can grow without any obstruction and harm the native population by taking the form of a weed (Kendrick 1992). Thus, weeds can cause problems not only to the crop yield and quality, but also to the nature by invading native environment and its species composition. In order to ensure the balance of ecosystems, global crop production and food security; weed management is an important factor to be considered (Bjawa 2014).

During the early 1990s weed management was dominated by mechanical methods (Sommer 1996). The importance of mechanical weed control has become limited as it causes soil erosion and nutrient losses (Zimdahl 1993). Thus, chemical herbicides became the viable alternative. However, due to continuous use of the same herbicide with the same mode of action, herbicide resistance increased and for the last 20 years no chemical has been manufactured with a different mode of action than the previous products (Duke 2012). Researchers also identified the herbicidal toxic impact on non-target soil organisms which play an important role in degrading and decomposing organic matter (Subhani et al. 2015; Zain et al. 2013). Therefore, the need for biological weed control as well as development of bio-herbicides has become important.

Biological control of weeds is an effective and efficient alternative control method (Senthilkumar 2007). Biological weed control is used against invading plant species that pose a threat to endangered ecosystems and is designed to reduce the competition for nutrients and space by weeds. The aim of biological control of weeds is to decrease, suppress or kill the weed population using host specific insect herbivores or plant pathogens such as fungi, bacteria and viruses (Bailey et al. 2010).

The initial concept for using fungal pathogens in weed control was observed by a farmer in 1890 in the USA with the thistle population controlled by a rust fungus (Wilson 1969). A similar situation was recorded by Morris (1991) who observed that the Australian gall-forming rust fungus Uromycladium tepperianum helps to reduce the infestation of Acacia saligna at over 50 localities in South Africa. Until 1970 biocontrol of weeds was not considered as a solution or alternative to the massive use of chemical herbicides (Pointing and Hyde 2001).

Biological control refers to the planned introduction of an exotic bio control agent for permanent establishment for long term control in an area where weeds are problematic to the natural habitat (Evans 1998; Eilenberg et al. 2001). Fair examples for using fungi in biocontrol of weed are the successful use of the European rust fungus Phragmidium violaceum to control European blackberry (Rubus sp.) in Chile, or the use of Puccinia chondrillina to control Chondrilla juncea (rush skeleton weed) in Australia, which is considered as the most remarkable successes ever achieved with biocontrol (Kendrick 1992). These rust fungi are obligate biotrophs and cannot mass produce spores (or even grow) in an artificial medium. Therefore, small amounts of natural inoculum were introduced to the area and the weed control was obtained by natural spore production and dispersal (Cullen et al. 1973; Cullen and Delfosse 1985; Kendrick 1992; Espiau et al. 1998).

Inocula of plant pathogens, applied to weeds in a similar manner to synthetic herbicides are called bioherbicides. When fungi are involved, they are referred to as mycoherbicides (Boyetchko et al. 1996). These pathogens usually occur naturally on weeds in localities where they need to be controlled, but not in sufficient amounts. Therefore, fungal inocula are mass produced and sprayed on to the weeds. These mycoherbicide products must undergo a government regulated registration process similar to chemical herbicides before they are released for public use (Evans 1998).

Mycoherbicide studies began in the 1940s in several countries with the aim to spread indigenous pathogenic fungal species on target weeds as a control measure (Pointing and Hyde 2001). In 1949, an attempt to control prickly pear cactus using Fusarium oxysporum was unsuccessful (Julien and Griffiths 1998). Kendrick (1992) mentioned another few examples where mass produced fungal propagules are applied as a mycoherbicides. A Colletotrichum gloeosporioides spore suspension was used to control northern joint vetch (Aeschynomene virginica) in rice and soybean fields in the USA. This was the first practical use of a fungus as a mycoherbicide that was later commercialised as “Collego”. Another example was water hyacinth, considered to be the worst aquatic weed in the world, but could be controlled by Acremonium zonatum and Cercospora rodmanii. However, during the 1950s, spores of Alternaria cuscutacidae were mass produced and applied to a dodder (Cuscuta sp.) (Wilson 1969, Julien and Griffiths. 1998). In 1963 China developed a mycoherbicide for dodder, using Colletotrichum gloeosporioides cf. cuscutae (Colletotrichum cuscutae) (Auld 1997; Kendrick 1992).

In the 5th edition of “A world catalogue of agents and their target weeds” by Winston et al. (2014), seven fungal species are mentioned that were developed as mycoherbicides and some of these are available as commercial products. According to Walker and Connick (1983) and Auld (1993) dew and temperature are the most important factors for primary infection of pathogen in order to obtain a successful disease development on the target weed. Some researchers have insisted that it is important to consider the time for secondary infection; which lead to a successive distribution of disease in the field (Boyette et al. (1979). According to Hasan and Ayres (1990) and El Morsy (2004), Stagnospora species on Calystegia sepium required only 3 weeks for control and Alternaria alternata on water hyacinth took 2 months.

Finding a new active fungal isolate for mycoherbicide production is not an easy task. There are a few important factors that need to be considered before mass production (O’Connell and Zoschke 1996). Mass availability of product, scientific testing for laboratory and field conditions, registration and commercializing procedures are necessary (Evans 1998). Within the past three decades research for improving this technology has increased. Even though the target weeds and phyto-pathogenic species identifications are still ongoing, interesting discoveries have surfaced overtime. According to Gan et al. (2013), the number of candidate genes found from the genomes of Colletotrichum gloeosporioides and Colletotrichum orbiculare, which were predicted to be involved in pathogenesis, showed the potential to be explotied as mycoherbcides. The discovery of genes in these two organisms related to the production of Indole acetic acid (IAA), a component of some of the well established herbicides, showed that these can be converted to mycoherbicides (Gan et al. 2013). Some Phoma species were also considered as a successful candidate for the biocontrol of weeds. The ability of Phoma macrostoma to inhibit the growth of dicot plants was studied (Bailey et al. 2011, 2013; Smith et al. 2015). This fungus was used to control broadleaf weeds in turf systems in Canada and the USA. A registered commercial product of Phoma macrostoma is also available in Canada and USA (Evans et al. 2013).

The secondary metabolites produced by some fungi have been shown to have herbicidal activity. Castro de Souza et al. (2016) found several Diaporthe spp. from the Brazilian Pampa biome that have the ability to produce secondary metabolites with herbicidal activity. The genus Diaporthe is very rich in secondary metabolites, as recently summarised by Chepkirui and Stadler (2017).

Biological control may take several years to take effect and the effectiveness is influenced by a number of factors, such as climatic conditions, geographical region and management practices (Pointing and Hyde 2001). Among these, high initial cost, limited number of natural enemies and uncontrollable dissemination of biological control agents after its release in nature are considered as disadvantages (El-Sayed 2005). Biological control is particularly useful in areas where other conventional control methods are inappropriate, uneconomic or unachievable (Reznik 1996).

Different fungal species act as a promising source for the production of various compounds that can be used as potential herbicides. Since many of these toxins play a key role in the development of plant diseases (Pointing and Hyde 2001); the potential of these chemicals as herbicides can also be explored. When it comes to controlling weeds, herbicide-resistant weeds can be a challenge for conventional control methods. Therefore, there is a potential to find compounds that can act as models for developing herbicides with new modes of action (Castro de Souza et al. 2016). Through in depth studies on the potential of fungi and their products, more environmentally friendly herbicides can be produced for sustainable and eco-friendly control of weeds.