key: cord-0947969-1vz26ser
authors: Cock, Ian E.; Van Vuuren, Sandy F.
title: The traditional use of southern African medicinal plants in the treatment of viral respiratory diseases: A review of the ethnobotany and scientific evaluations
date: 2020-07-27
journal: J Ethnopharmacol
DOI: 10.1016/j.jep.2020.113194
sha: 431dbae5c314072152a003842617253e461ee9ac
doc_id: 947969
cord_uid: 1vz26ser

ETHNOPHARMACOLOGICAL RELEVANCE: Viral respiratory infections are amongst the most common infections globally, with most of the world's population contracting at least one infection annually. Numerous plant species are used in traditional southern African healing systems to treat these diseases and to alleviate the symptoms. Despite this, the therapeutic potential of these plants against viral respiratory diseases remains poorly explored. AIM OF THE STUDY: The aim of this study was to document the southern African plant species used in traditional medicine to treat viral respiratory infections. We also examined the extent of scientific evaluations of southern African plant species against the respiratory-infective viruses, with the aim of stimulating interest in this area and focusing on future studies. MATERIALS AND METHODS: We undertook an extensive review of ethnobotanical books, reviews and primary scientific studies to identify southern African plants which are used in traditional southern African medicine to treat viral respiratory diseases. This information was used to identify gaps in the current research that require further study. RESULTS: Two hundred and fifty-seven southern African plant species were identified as traditional therapies for viral respiratory diseases. Surprisingly, only one of those species (as well as twenty-one other species not recorded for these purposes) has been evaluated for the ability to block respiratory virus production. Furthermore, most of these studies screened against a single viral strain and none of those studies examined the mechanism of action of the plant preparations. CONCLUSIONS: Despite well documented records of the use of southern African plants to treat respiratory viral diseases, the field is poorly explored. Nearly all of the plant species used in traditional healing systems to treat these diseases are yet to be tested. Substantial further work is required to verify the efficacy of these traditional medicines.

Respiratory diseases are amongst the most prevalent diseases globally and impose an immense burden on the healthcare system. Several of these diseases are common causes of severe illness and mortality (Wang et al., 2015) . The greatest burden of respiratory diseases is due to pathogenic infections. Acute lower respiratory tract infections are one of the top causes of mortality and disability globally, causing approximately four million deaths annually . These conditions are particularly concerning in children and are the leading cause of mortality in children under five years of age. Influenza virus infections account for many acute lower respiratory tract infections and it is estimated that 250,000-500,000 people die from severe influenza virus infections annually. While Semenya and Maroyi (2018) reported on 306 medicinal plant species used to treat and manage respiratory infections, a concerted effort to date has not been undertaken in catagorising these medicinal plants specifically to respiratory viral and bacterial sources. This review focusing on southern African medicinal plants for viral infections forms part one of a two part review whereby the focus on bacterial pathogens is reviewed as Part 2 in a separate manuscript (Cock and Van Vuuren, 2020) .

Several viruses can infect respiratory tissue and induce respiratory illnesses. Of these, the most common viral causes of respiratory disease are influenza virus, rhinoviruses, respiratory syncytial virus (RSV), parainfluenza virus, adenovirus, human metapneumovirus, and enterovirus.

Influenza (often abbreviated to flu) is a highly infectious airborne disease caused by the influenza virus. There are four genus's of influenza virus, of which three genus's (A, B and C) can infect humans (Paules and Subbarao, 2017) . Influenza virus D genus has never been reported as a human infection and instead only infects pigs and cattle. However, it is believed that this genus has the potential to infect humans and cross species infections may be reported in the future (Asha and Kumar, 2019) . Of the other three classes of influenza virus, influenza virus A is the most important to human health and is the most virulent of the human influenza pathogens (Paules and Subbarao, 2017) . Whilst this genus consists of only a single species, numerous serotypes have been described (listed in descending order of deaths caused):

• H1N1. This serotype has several viral sub-serotypes which have caused considerable mortality. The H1N1 serotypes were responsible for the Spanish influenza outbreak of 1918 (which infected approximately 500 million people worldwide and caused an estimated 50-100 million deaths) (Taubenberger and Morens, 2006) , and the swine influenza outbreak of 2009 (which infected approximately 20% of the world's population and caused thousands of deaths) (Reddy et al., 2018) .

• H2N2. This serotype was the cause of the Asian influenza outbreak of 1957. This serotype may have resulted from a mutation in ducks that subsequently crossed species to humans (Scholtissek et al., 1978) . The death toll of this virus has been estimated to be between one and four million.

• H3N2 serotype caused the Hong Kong influenza outbreak of 1968, which is estimated to have killed at least a million people between 1968 and 1969. This serotype is believed to have descended from H2N2 via an antigenic shift mutation (Scholtissek et al., 1978) .

• H5N1, also known as avian influenza or bird flu. Whilst this virus can infect humans and several other animal species, its virulence in humans is relatively benign compared to the previously described serotypes. This virus is highly pathogenic in many bird species and an outbreak of bird-adapted strain of H5N1 in 2004 in southeast Asia directly resulted in the deaths of tens of millions of birds. Hundreds of millions of more birds (particularly domestic fowl) were culled to inhibit the spread of the disease, causing serious economic hardship in the region. Airborne spread of this sub-serotype virus from birds to humans is not yet possible, although it is believed from genetic analyses that only a few mutations are required for this to occur (Sorrell et al., 2011) , raising concerns of a future pandemic.

• H7N7 is a zoonotic serotype that can infect humans, pigs, horses, seals and birds by natural transmission routes in the wild, and has been used to infect mice in laboratory studies (Paules and Subbarao, 2017) . The zoonotic nature of the virus indicates that is poses a future pandemic threat.

• H1N2 is also a zoonotic serotype, which infects humans, pigs and birds (Paules and Subbarao, 2017) . Due to its cross-species infectivity, it is also believed to be a future pandemic threat.

• H9N2, H7N2, H7N3 and H10N7 are other bird-infective serotypes, although they can also infect humans, especially children (Paules and Subbarao, 2017) .

A common feature of the influenza virus A serotypes, whether they infect humans or not, is their high rate of mutation and therefore their potential for major pandemics. If such pandemics occur, lack of immunity, ease of transmission and increased virulence may cause major losses of life, as previously seen for the Spanish influenza outbreak over a 100 years ago. It is therefore important that therapies to treat these infections be developed to guard against future pandemics.

Influenza virus genus's B and C are less common than influenza virus A and are generally considered to be less of a concern due to their lower virulence and slower mutation rates (Paules and Subbarao, 2017) . Indeed, influenza virus B mutates at a third to a half of the rate of influenza A. Furthermore, this virus only infects humans. The reduced rate of antigenic change, combined with the single species host infectivity (thereby inhibiting crossspecies antigenic shifts), substantially decreases the chances of pandemic shifts occurring in these influenza genus's.

There are currently relatively few effective cures for any influenza virus species. The exception are the neuraminidase inhibitors which prevent influenza virus reproduction by inhibiting viral budding from the cell. Most other therapies target the symptoms of the illness to make the infected person feel better whilst their immune system fights the infection.

However, influenza vaccines are now available against most influenza serotypes, although the vaccine effectiveness ranges from 10-60% (depending on the year and the serotypes targeted by the vaccine) (Osterholm et al., 2012) . Therefore, many people will still contract influenza, despite receiving a vaccine and an effective therapeutic alternative is urgently required.

However, these infections can also cause substantially more serious symptoms including inflammation of the heart and surrounding tissue , inflammation of the membranes surrounding the brain and inflammation of the pancreas. There are currently no cures for any enterovirus-induced diseases. Instead, treatment is supportive and aims to alleviate the symptoms (e.g. providing analgesics for pain relief) and more effective therapies are required. For the polioviruses, effective vaccines already exist, which allow for the prevention of poliomyelitis. However, these vaccines are ineffective if a poliovirus infection is already established. To date, there are no effective vaccines for the other respiratory infective enteroviruses.

infections. However, HPIV infections can be much more serious in younger children. HPIV-1 and HPIV-2 infections can induce severe upper respiratory tract distress in children up to four years of age. HPIV-3 can cause bronchitis and viral pneumonia in infants (<1 year of age).

Similarly, HPIV can cause severe pneumonia that can be fatal in immunocompromised people. Therefore, although HPIV research is relatively neglected, effective anti-HPIV therapies are required.

Human respiratory syncytial virus (HRSV) is another common cause of respiratory virus infections, particularly in infants and children. Infections are very common and it is estimated that 60% of infants are infected in their first year and nearly all children are infected by 3 years of age (Glezen et al., 1986) . In most infected people, HRSV infections are asymptomatic, or produce relatively benign symptoms, similar to a common cold. However, HRSV infections may induce severe bronchitis requiring hospitalisation in approximately 3% of the population and prior HRSV infections do not confer lifelong immunity (Hall et al., 2009 ). Instead, people may acquire repeated infections as early as 1 year after a prior infection. There is no cure for HRSV infections and treatment is supportive and aims to alleviate the disease symptoms and allow the infected persons immune response to combat the disease. In older people with a decreased immune response and other immunocompromised individuals, HRSV infections may be substantially more severe and in rare cases may result in death.

HRSV is very readily spread and avoiding infection is almost impossible (Glezen et al., 1986) . Unlike other respiratory viruses where the route is mostly airborne, HRSV transmission is most frequently via direct contact. The virus is particularly hardy and remains viable on surfaces for up to five hours. The virus is also relatively resistant to chemical decontaminants, making its prevention particularly difficult. For this reason, HRSV spreads rapidly through childcare centres and treatment facilities for the elderly. HRSV infections are most frequent in the colder months for most regions of the world, when people congregate together indoors, with closer contact and thus more efficient transmission.

HMPV infections are common causes of respiratory illness and are associated with 6-40% of viral respiratory infections in children seeking medical intervention. Indeed, serological studies indicate that nearly all children worldwide have been exposed to HMPV by 5 years of age (Kahn, 2006) . However, as for HRSV, prior infections with HMPV do not provide life-long immunity and repeated infections are common in older children and in adults. Generally, HPMV infections result in mild upper respiratory tract infections and present with similar symptoms to the common cold. However, the symptoms may be substantially more severe in older people and immunocompromised individuals and whilst relatively rare, HMPV-associated fatalities may occur. HPMV infections may also produce substantially more severe symptoms in individuals with asthma or chronic obstructive pulmonary disease (COPD).

The method of transmission of HMPV is less understood than for the other respiratory-infective viruses although it is likely to spread via similar pathways (i.e. airborne transmission, direct contact) (Chow et al., 2016) . There is currently no known cure for HMPV infections and there is relatively little research into developing a treatment as the disease is relatively benign in most people and is self-limiting, usually resolving in similar time as a common cold. However, ribavirin has shown efficacy in animal models and may be useful in severe HPMV infections (Kitanovski et al., 2013) . A number of studies have also investigated the potential for a HPMV vaccine (Karron et al., 2017) although it is likely that this is still at least several years from development.

Coronaviruses also cause respiratory tract infection that are usually mild compared to many of the previously listed viruses. However, rarer coronaviruses such as the viruses those that cause severe acute respiratory syndrome (SARS) and Middle East respiratory syndrome (MERS) induce substantially more severe symptoms and can even be lethal. Lastly and more topically and important is the SARS-CoV-2 coronavirus serotype which has caused recent global panic. The identified 2019-nCoV genome has been sequenced to some betacoronaviruses detected in bats, with initial zoonotic mutations from snake expected (Ji et al., 2020; She et al., 2020) . This remains at present the number one global viral respiratory threat.

To date, neither SARS nor MERS have yet had a significant impact on southern Africa and therefore are not a focus of this study. However, recent SARS-CoV-2 infection and mortality rates in South Africa have highlighted the global nature of COVID 19. Indeed, as at 13 May 2020, the number of infections in South Africa was in excess of 12,000 and the mortality rate was above 200 (worldometersiinfo/coronavirus/). This figure is highly dynamic and is likely to be substantially higher by the time this review is published. Furthermore, due to the relatively high number of asymptomatic individuals with SARS-CoV-2 infections, these figures are likely to be substantially underestimated. The ease of transmission of this virus highlights the need to test southern African plants against this serotype in future studies.

It is noteworthy that all the respiratory viruses discussed here (with the exception of adenoviruses) are RNA viruses and this may impact on the efficacy of any prevention or therapy in the future. In general, RNA viruses incorporate mutations at a substantially faster rate than for DNA viruses. Indeed, the mutation rates in influenza A virus has been estimated to be approximately one mutation per genome per replication (Drake, 1993) . This mutation rate is up to 300 times faster than reported for DNA viruses. It is likely that this high mutation rate may induce changes to viral surface proteins. As antibodies are produced against these proteins following prior exposure (either via exposure to the virus or through vaccination), this may impact on an individual's immunity to these viruses. This is particularly apparent for influenza viruses and annual influenza vaccinations include updated antigens to account for this and to provide for immunity to new and emerging strains. The mutability of the other RNA respiratory viruses also affects an individual's ability to respond to future reinfections. Indeed, it was noted in the sections discussing the individual respiratory viruses that prior exposure generally does not confer life-long immunity.

Similarly, the mutability of the RNA viruses may provide them with resistance to antiviral drugs and the new strains should be screened against these drugs as those strains. This also highlights the need to develop new antiviral therapies.

Also noteworthy, influenza viruses, parainfluenza viruses, HRSV, HMPV and coronaviruses are enveloped viruses, whereas the viruses that cause colds (enteroviruses, rhinoviruses, adenoviruses) are non-enveloped viruses. Non-enveloped viruses have substantially greater stability than enveloped viruses outside of the host cell as lipid envelopes are relatively fragile and sensitive to denaturation by environmental factors (e.g. temperature and pH) and well as by detergents and other chemicals (Lucas, 2001) . For this reason, using detergents and soaps to wash hands is considered an effective preventative measure for enveloped viruses including influenza virus and is recommended for the current SARS-CoV-2 pandemic. In contrast, the viruses that cause colds are non-enveloped and therefore more stable outside host cells. For this reason, they last for longer times, which accounts for the greater rate of spread of colds.

Viral respiratory infections are very common, largely due to their ease of transmission. Viral respiratory infections may be spread by inhalation when an infected individual coughs or sneezes, or even when they talk. Small droplets of saliva containing large numbers of the viral pathogen are released and dispersed into the air. If other individuals in the vicinity of the sick person breathe in these droplets, and they may become infected. Unless that person has already acquired immunity to the virus via prior exposure, it is likely that they will develop the illness. Thus, one of the most effective ways to retard the spread of respiratory viruses is via behaviour modification. If an infected person covers their mouth when coughing or sneezing, the incidence of spread of the viral respiratory disease is dramatically reduced. Respiratory viruses may also be spread via direct contact with an infected person, or via indirect contact with items and surfaces onto which an infected person has coughed or sneezed. Frequent and effective use of hand washes and surface disinfectants may therefore also substantially reduce the risks of infection. As adenoviruses and enteroviruses may also infect the gastrointestinal tract, contact with objects and surfaces contaminated with faeces from an infected individual (particularly via the hands) may also result in respiratory infections. Disinfection of hands and surfaces also reduces the spread of these viruses.

Whilst a number of different viral pathogens may cause respiratory disease, there is similarity between the symptoms of many of those infections. Coughing, sneezing, runny noses, sore throats, blocked airways and laboured breathing are common symptoms of viral respiratory infections. Fever may also be associated with some infections (e.g. influenza).

Infected individuals may also suffer from headaches and/or muscle aches and pains, and lethargy is common. The onset of the symptoms begins within a day or two of contracting the virus and last for up to 14 days (or longer for influenza).

In addition to differences related to the viral pathogen, the site of the infection (i.e. upper or lower respiratory tract infections) also significantly affects the disease symptoms.

Upper respiratory tract infections affect the upper respiratory passages (larynx, nasal cavity and nasal passages, as well as the pharynx). The common cold is an example of an upper respiratory infection. In addition to the symptoms listed above that are common for all viral respiratory infections, upper respiratory infections are also characterised by discomfort of the nasal passages, excess mucus production, nasal congestion, runny nose and a sore throat.

Headache, muscular pain and fever may also occur.

Lower respiratory tract infections occur when the virus infects the bronchial tubes (causing bronchitis) or the alveoli (causing pneumonia). The severity of the symptoms is dependent on the infective virus and the severity of the infection. Less severe infections may present with similar symptoms to the common cold (described above). In more severe infections, a phlegm-producing cough, fever, difficulty breathing, chest pain and wheezing may also be evident. Many of the symptoms are common between upper and lower respiratory tract infections. However, the nature of the symptoms can help differentiate between the sites of infection. People with lower respiratory infections generally present with consistent coughing with high phlegm production as the most apparent symptom. For upper respiratory tract infections, sneezing, sore throats and headaches are more apparent.

There are few effective treatments for viral respiratory diseases and medical intervention is generally not required for most individuals. Viral respiratory diseases are generally self-limiting and most people will recover with rest and the consumption of adequate volumes of fluids. In most cases, therapy aims to alleviate the symptoms rather than to treat the cause of the disease. Acetaminophen (paracetamol) is useful for relieving much of the pain and discomfort associated with viral respiratory infections. Antiviral medicines are usually not used except in severe and/or prolonged cases of influenza. As of 2020, six prescription influenza antiviral medications had been approved by the United States Food and Drug Administration (FDA):

• Three chemically and functionally related drugs: oseltamivir phosphate (known as Tamiflu®), zanamivir (marketed as Relenza®) and intravenous peramivir (Rapivab®), are used as pharmaceutical treatments of influenza (FDA, 2019). All of these medications block the viral neuraminidase enzymes, thereby inhibiting viral release from infected cells. These neuraminidase inhibitors are effective against both influenza A and B viruses. They are relatively effective and generally are safe for most people to use, although side effects including nausea and vomiting have been reported in some people. The cost of these drugs can be prohibitive in some regions of the world (particularly in developing countries) and they are often not available in rural and isolated regions. Furthermore, influenza virus resistance is increasingly being reported to these drugs (Sheu et al., 2008) and new therapies are urgently required.

• Baloxavir marboxil (marketed as Xofluza®), is also effective for the treatment of influenza A and B infections (FDA, 2019) . This drug functions by inhibiting the cap-dependent endonuclease activity of influenza viruses, thereby blocking influenza virus RNA synthesis.

However, several influenza strains have rapidly developed resistance to this drug since its clinical introduction in 2018, already limiting its efficacy (Hayden et al., 2018) . Furthermore, adverse side effects including bronchitis and diarrhoea are reported in >20% of people who take this drug.

• The adamantine drugs amantadine (Symmetrel®) and rimantadine (Flumadine®) target the M2 ion channel protein of influenza A viruses (FDA, 2019) . Therefore, they are indicated for influenza A but are ineffective against influenza B viruses. However, several influenza A serotypes have developed high levels of resistance (>99%) to these drugs and they are no longer considered effective against H1N1 and H3N2 serotypes (Deyde et al., 2007) .

Whilst influenza virus resistance to the neuraminidase and cap-dependent endonuclease inhibitors classes of drugs are still relatively low, this is likely to change in the future with continued overuse of these drugs. These drugs should be used sparingly and reserved for high risk individuals (children, the elderly and immunocompromised individuals). Furthermore, medical practitioners will also often prescribe antibiotics for viral respiratory infections. Antibiotics have no effects against viruses and they should not be used for viral respiratory diseases unless a secondary bacterial infection is suspected. The incorrect use of antibiotics for these purposes may result in antibiotic resistant bacteria and it is recommended that this practice should cease.

Plant-based and traditional medicines are also used by many cultures to treat viral respiratory infections and their symptoms. This is especially true in developing countries. In many areas (particularly in isolated and rural communities), consultation with traditional healers is prevalent and is often the primary healthcare modality. Furthermore, allopathic drugs are often perceived in these areas as being relatively ineffective and are generally expensive compared to traditional medicines. Isolated rural communities may have limited access to allopathic medical practitioners and conventional pharmaceuticals are often not readily available. An examination of the antiviral properties of traditional medicines against viral respiratory pathogens may highlight new leads for drug discovery and may provide effective new therapies to treat viral respiratory diseases.

Southern Africa has similar viral respiratory disease epidemiology trends to other regions with similar population densities. The incidence of viral respiratory diseases substantially increases in the colder winter months (commonly referred to as cold and flu season) (Gessner et al., 2011) . Geographical differences for viral respiratory diseases are also evident in various parts of southern Africa. Viral respiratory diseases are substantially more frequent in cold mountainous regions. These regions tend to have lower temperatures and therefore people tend to congregate indoors for longer periods, facilitating viral transmission.

Related ethnic correlations also occur. For example, the southern Sotho tend to have higher prevalence's of the viral respiratory diseases than other ethnic groups in South Africa. The increased prevalence is due to the living environments of this group, rather than genetic factors. As the Southern Sotho often live in cold mountainous regions, they spend more time indoors in groups than many other ethnic groups. Thus, viral transmission rates are higher.

Incidentally, the higher prevalence of the viral respiratory diseases in the southern Sotho correlates with the development and usage of substantially more therapies for these diseases than have been reported for other ethnic groups. Indeed, nearly three times the amount of plant species have been recorded for the treatment of respiratory virus diseases by the southern Sotho compared to the Zulu (as discussed later and summarised in Table 1 ). This is despite the population size differences between these groups and is in contrast to the substantially higher volume of studies reported for the Zulu ethnobotany. Indeed, Hutchings et al. (1997) and Ngwenya et al. (2003) are concerned only with Zulu ethnobotany, whereas many of the other texts used in this study examine the ethnobotany of multiple groups (Smith, 1888; Watt and Breyer-Brandwijk, 1962; Von Koenen, 2001; Van Wyk et al., 2009) .

Similarly, a large proportion of the ethnobotanical surveys cited in our study also concentrate on Zulus and regions in Zulu areas. In contrast, a single book (Moffett, 2010) and limited ethnobotanical surveys (Kose et al., 2015; Moteetee, et al., 2019) have focused specifically on southern Sotho ethnobotany.

High density lower socio-economic urban communities often have higher prevalence's and rates of transmission of viral respiratory infections than other populations.

Living in high density conditions facilitates viral transmission rates. Rural communities may also have different disease profiles, with higher incidences of zoonotic viral strains than in urban populations. Some respiratory viruses can infect domestic animals and livestock, as well as humans. For example, then H1N1 swine influenza strain, as well as H7N7 and H1N2 serotypes infect both humans and pigs. The latter two serotypes also infect birds and other farm livestock. All of these viruses are readily infective by airborne transmission between species. The H5N1 serotype (avian flu) can also infect both humans and birds, but is not yet readily transmissible to humans by airborne routes. People living in rural communities may have close contact with livestock, providing an extra avenue for infection. Therefore, these serotypes are often more common in rural communities than in urban environments. Several other attributes are shared between lower socio-economic urban communities and rural communities. For example:

•both communities often have limited access to effective clinical treatment. Medical care is often not affordable to poor and/or rural people. Instead, these communities tend to be more reliant on traditional medicine. Alternatively, given the comparatively low concern about many of the viral respiratory diseases compared to other infectious pathogens, and their selflimiting nature, many people suffering from these illnesses forgo medical treatment and allow the infection to run their course.

• Economically disadvantaged people and those living in rural communities are often less educated and have less understanding of simple preventative measures (e.g. frequent hand washing, disinfecting surfaces etc.) that limit the transmission of viral respiratory diseases.

The prevalence and severity of symptoms of viral respiratory diseases often correlates with age. Indeed, several of these diseases are generally considered to be childhood diseases only. A very common childhood disease is HRSV and nearly 100% of children have contracted HRSV by the age of three (Glezen et al., 1986) . Whilst prior contact does not provide lifelong immunity to this disease, it does provide some protection and adult HRSV infections are far less common. Parainfluenza in also substantially more common in children than in adults as most children contract a HPIV infection before the age of 10 years (Henrickson et al., 1994) , conferring them a degree of immunity. In most children, this is a relatively innocuous disease, although the symptoms may be substantially more serious in infants and can be life threatening. Children are also more prone to influenza and colds than adults because of their relative lack of immunity to these diseases. Throughout the course of their life, humans encounter multiple respiratory viral pathogens. In response, we develop antibodies against these viruses which confer protection against future infections with the same virus. Thus, adults often only contract influenza infections every few years as they encounter influenza strains they have previously not contracted. The large number of viral pathogens that cause colds allows adults to potentially contract multiple colds within a single year. However, once a particular strain has been encountered, the immune system confers protection against future infections with that specific virus. In contrast, children will usually have encountered relatively few respiratory viruses and therefore have limited immunological protection so they may be infected far more frequently than adults.

The incidence and severity of viral respiratory infections is also higher and the symptoms more severe in older people. The immune systems of the elderly are less efficient than in the younger generation, allowing these diseases to last for longer periods and allowing the symptoms to be substantially more severe. Indeed, influenza and colds are of considerable concern in elderly people and the risk of mortality is substantially increased. Similarly, immunocompromised individuals are also at far greater risk of contracting a viral respiratory disease than the general population, and when they contract these diseases, the symptoms are often substantially more severe and may be life threatening (Nunes et al., 2014; Annamalay et al., 2016) . This is particularly concerning for southern Africa. Sub-Saharan Africa, including South Africa, has a substantially higher prevalence of HIV infected people than many other regions of the world. Indeed, as of 2018, it was estimated that 7.52 million South Africans (equating to >13% of the entire population and 19% of the adult population) are living with HIV (Statistics South Africa, 2018). This is particularly concerning in a southern African context as these individuals with untreated HIV would have be expected to have greatly increased rates of respiratory virus infections, and the infections have far greater consequences. Timely diagnosis and the usage of highly active antiretroviral therapies (HAART) is necessary to overcome these more profound effects in individuals with HIV.

Vaccinations to some of the respiratory infective viruses are readily available. In particular, influenza vaccinations are updated annually to provide protection against the latest influenza strains. Their efficacy has been improving in recent years, now consistently providing protection rates of 50-60%. These vaccinations are made readily available. Despite this, the uptake of these services is particularly low in South Africa. Indeed, it was recently estimated that <5% of South Africans received the influenza vaccination in 2015 (Solanki et al., 2018) . As vaccination is currently the best preventative option for influenza, the rate of uptake is low and higher risk (children, the elderly and immunocompromised) people in particular would benefit from annual vaccinations.

This study aimed to identify southern African plants which are used in traditional healing systems for the treatment of viral respiratory diseases in humans. This information was sourced from a variety ethnobotanical books (Smith, 1888; Watt and Breyer-Brandwijk, 1962 ; Van Hutchings et al., 1996; Von Koenen, 2001; Ngwenya et al., 2003; Wyk et al.., 2009) , as well as ethnobotanical reviews (Hulley and Van Wyk, 2017; De Beer and Van Wyk, 2011; Nortje and Van Wyk, 2011; Philander, 2011; Van Wyk, 2008) . Google-Scholar, Science-Direct, PubMed and Scopus electronic databases were searched to identify original ethnobotanical research papers using the following terms as filters, and were searched both alone and as combinations: "South African", "medicinal plant", "traditional medicine", "ethnobotany", "respiratory infection" "colds", "influenza", "bronchitis", "respiratory syncytial virus", "RSV", "parainfluenza", "rhinovirus", "human metapneumonovirus". These terms were searched alone and as combinations. Only English language publications were used in the preparation of this review. This study is non-biased and does not have taxonomic preference (although several of the studies we review targeted specific families or genera).

All species names were checked using the Plant List website (http://www.theplantlist.org/).

Whilst most of the plant species included herein are native to southern Africa, introduced plant species were also included where there is evidence of their usage in at least one southern African traditional healing system. Our initial literature search identified two hundred and sixty-eight plant species that have been used as traditional remedies for viral respiratory diseases in southern Africa. The vast majority of these species are native to southern Africa, although, several introduced species were also included. For introduced species to be included in this report, they must either be naturalised or widely cultivated, and there must be documented evidence that they are commonly used by at least one southern African ethnic group to treat viral respiratory disease. these studies were excluded from this review unless they specifically mention the use of the plant species to also treat human respiratory viral infections. We aimed through this review to summarise traditional southern African ethnopharmacological knowledge on the treatment of viral respiratory diseases, and to discuss the plant species used traditionally to treat these infections. Through this study, we aim to foster further research into southern African plants as treatments for viral respiratory diseases by highlighting plant species traditionally used to treat those diseases.

Viral respiratory diseases exhibit several symptoms that are relatively generic and also occur with multiple other diseases. Traditional medicine practitioners often use plant preparations to target disease symptoms rather than specifically targeting the disease. Indeed, many ethnobotanical texts and reviews record the use of plants to treat symptoms such as coughs and runny noses. Many other diseases produce similar symptoms. For example, other bacterial respiratory diseases such as tuberculosis elicits similar symptoms. Likewise, the early phases of acquired immunodeficiency syndrome (AIDS), anthrax, bubonic plague, cytomeglavirus, Lyme disease, malaria, measles, rabies, severe acute respiratory syndrome (SARS) and smallpox (to list just a few) present with similar symptoms to the viral respiratory diseases and may be mistaken for influenza. Other ethnobotanical texts may list plants used to treat "chest complaints", 'breathing disorders" or other similar symptoms. In this study, only plant species that have specifically been reported to be used to treat viral respiratory infections are included. Where ambiguity occurs regarding a plants specific usage, that plant has been excluded from this study.

An extensive literature search identified 257 southern African plants used in at least one southern African traditional healing system to treat viral respiratory infections (Table 1) .

The high number of species used to treat these diseases is hardly surprising given the prevalence of these infections. Indeed, it is likely that this list is incomplete and has Numerous ethnobotanical surveys used in the preparation of Table 1 also reports on medicinal plant use by ethnic groups in specific locations (Hulley and Van Wyk, 2017; De Beer and Van Wyk, 2011; Nortje and Van Wyk, 2011; Philander, 2011; Van Wyk, 2008) .

Together, these studies provide an evaluation of the usage of southern African plants for the treatment of viral respiratory disease and allow for the detection of specific trends.

Interestingly, a high proportion of the studies which mention the usage of plants by specific ethnic groups to treat viral respiratory diseases relate to southern Sotho traditional medicine. Indeed, of the plant species that were listed as being used by specific ethnic groups, 70 species were listed as used by the southern Sotho, 26 species are listed as used by the Zulu, 16 species by the Xhosa, five species each by the Tswana and Nama, four species by the San, and one species each by the Venda and Khoikhoi. This is hardly surprising given the environments that each of these groups live in. The southern Sotho live at higher altitudes and in substantially cooler environments than the other ethnic groupings. These conditions require the Sothern Sotho to spend greater periods of time indoors, living in close proximity, especially during the colder periods. Such conditions are ideal for the transmission of respiratory viruses and it is likely that the southern Sotho would contract these infections at higher rates than the groups living in warmer environments.

It is also noteworthy that the Southern Sotho has substantially more treatments for the common cold than they did for influenza. This is also likely to correspond to the transmission of these viruses. More than 200 viral serotypes (including over 160 rhinovirus serotypes) have been identified as causes of colds. Due to the antigenic variability of these viruses, otherwise healthy adults may catch up to two or three colds a year. In contrast, there are less influenza virus serotypes, although frequent mutations provide new antigenic variants.

However, on average, otherwise healthy adults will only contract an influenza infection every five years. Thus, although influenza infections are substantially more severe than colds, the frequency that the southern Sotho contracted each illness would make cold treatments of particular importance to them.

A further trend regarding the usage of the medicinal plants was also evident. Previous studies have reported that decoctions and infusions are the most common methods for treating most pathogenic diseases (Cock et al., 2018; Cock et al., 2019) . Whilst these methods of preparation and usage were also the most common for the treatment of respiratory viruses, burning the plant part and inhaling the smoke was substantially more common for this purpose than previously reported for the treatment of other diseases. This is not unexpected as burning of plant volatiles for pulmonary treatment regimens is common practice from a global perspective (Mohagheghzadeh et al., 2006) Indeed, of the 70 plant species specifically listed as used by the southern Sotho for the treatment of viral respiratory diseases, 21 of these were burned and the smoke inhaled. A further four of the southern Sotho medicinal plant species were burned and used to fumigate the huts of sick people. Inhaling the smoke from burning plant materials was less frequently used by other ethnic groups, although this treatment method was also used by the Zulu and Xhosa for one species each. A further four plant species were also smoked to treat viral respiratory disease, although the ethnic group was not specified for those species. Boiling the plant material and inhaling the steam was used for a further four plant species, although the ethnic group(s) using these species was only stipulated for one of these species (by the San). Several plant species were also powdered and used as a snuff. Five species used by the southern Sotho were administered in this way. The Zulu also used four plant species as snuffs, and a further two plant species (for which the ethnic group was not specified) were also used this way.

In some instances (five plant species) the fruit was indicated for treatment. It is possible that the vitamin C content was targeted here and hence symptomatic treatment was considered rather than a cure sought. and Xanthorrhoeaceae (Figure 1 ).

Few of these species have had their inhibitory activity validated against viral respiratory pathogens via in vitro testing ( Laminaceae families were particularly well represented for the treatment of fungal skin infections, with 55, 28 and 23 species identified from these families respectively (Figure 1 ).

The higher representation of these families compared to other families may indicate that they have better potential against respiratory viruses. However, these are large families, each consisting of many species. Indeed, Asteraceae consists of nearly 33,000 individual species and has a wide global distribution (Bremer et al., 1992) . Therefore, the greater representation of Asteraceae, Fabaceae and Laminaceae is consistent with the higher number of species in these families. Many of the plant species traditionally used for the treatment of viral respiratory infections are herbaceous. This is advantageous for sustainable production of these medicines as herbaceous plants grow rapidly. The usage of plant parts that does not extensively damage the plant is preferred, particularly for endangered and threatened plant species. Unfortunately, the ethnobotanical records were incomplete for many plant species identified as treatments for viral respiratory diseases and we were unable to identify which plant part is used. Further ethnobotanical studies are required to clarify this. 

Surprisingly few studies have examined the effects of southern African plants against viral respiratory infections (Table 2) . Indeed, we were only able to find nine studies that have reported activity for southern African plants for activity against respiratory viruses. Not surprisingly given its prevalence and more serious nature, the influenza virus has been the most extensively studied. However, we were only able to find five studies that reported antiinfluenza virus activity for some southern African plants (Lall et al., 2006; Michaelis et al., 2011; Rahmasaria et al., 2017; Mair et al., 2018; Mehrbod et al., 2018) The most extensive of these studies reported that five of the screened plants inhibited influenza virus A viral production against a H1N1 reference serotype (Mehrbod et al., 2018) . Similarly, a preparation called EPs 7630 prepared from P. sidoides inhibited the replication of H1N1and

H3N2 influenza virus, parainfluenza virus, HRSV and human coronavirus strain 229E at 100 µg/mL (Michaelis et al., 2011) , but did not affect H5N1 influenza strain or several adenoviruses and rhinoviruses. The susceptibility of the 229E coronavirus to this preparation is particularly interesting given the current SARS-CoV-2 pandemic and highlights this preparation as a potential treatment for COV-19. Whilst we were unable to find publications screening EPs 7630 against that virus, this would be a priority for future studies.

T. ventricosa was also particularly strong inhibitor of influenza virus replication, with an EC 50 of 0.05 µg/mL. This was notable inhibition and was similar to the inhibition recorded for the control antivirals amantadine hydrochloride and oseltamivir carboxylate. C. spicata and P. viridiflorum were also strong inhibitors of influenza virus A production, with EC 50 values as low as 5 and 3 µg/mL respectively. Noteworthy activity was also recorded for C.

glabrum and R. melanophloeos extracts, albeit with substantially higher EC 50 values (110 and 113 µg/mL respectively). However, T. ventricosa also displayed some toxicity, with a selectivity index (SI) of 2 reported. This may limit the use of T. ventricosa extracts, as only relatively low concentrations would be considered safe to use to treat influenza A. The lowest toxicity and highest SI (8) was determined for C. spicata. This plant may therefore be the preferred therapeutic option for the treatment of influenza A H1N1 serotypes.

Two other studies also reported influenza virus inhibitory activity for A. linearis (Rahmasari et al., 2017) and H. melanacme (Lall et al., 2006) . The Lall et al. study (2006) Three studies also screened southern African plant extracts against parainfluenza virus (Michaelis et al., 2011; Bagla et al., 2012; Kutluk et al., 2018) . The Kutluk et al. (2018) study screened H. armenium against parainfluenza and reported strong inhibition (MIC=4 µg/mL). The Bagla et al. (2012) study screened a more extensive panel of plant extracts and reported noteworthy activity for A. schimperi, C. edulis and E. capenses (Bagla et al., 2012) . It was perhaps surprising given the prevalence of colds, that we were only able to locate two study that screened southern African plant extracts against rhinovirus (Beuscher et al., 1994; Micaehelis et al., 2011) (Beuscher et al., 1994) . A single study also tested iridoid glycosides isolated from B.

prionitis against RSV and reported EC 50 values as low as 2.5 µg/mL for the pure compounds (Chen et al., 1998) . Unfortunately, that study did not test the crude extract against RSV and it is not known whether it also has noteworthy anti-RSV activity.

Interestingly, only a single species of plant (C. glabra) tested for the ability to inhibit the production of respiratory viruses corresponded to plants used in traditional healing systems to treat those infections (Mhlongo and van Wyk, 2019) . A further four species were not recorded as being used traditionally to treat these infections, although are closely related to other species used for this purpose. The inhibitory activity of H. armenium (Kutluk et al., 2018) and H. melanacme (Lall et al., 2006) were screened against parainfluenza and influenza virus A respectively. Whilst there is no record of these plant species being used as traditional medicines to treat respiratory viral infections, ten other Helichrysum species are recorded as being used for these purposes (Table 1) . Similarly, we found no evidence that

Carissa edulis was used traditionally to treat parainfluenza despite it being tested for this activity (Bagla et al., 2012) . However, the related species C. bispinosa is used to treat viral respiratory disease, although it is recorded for use against influenza, rather than parainfluenza infections (Mhlongo and van Wyk, 2019) . Similarly, we were unable to find evidence that P.

virgata was used to treat colds, despite it being screened against rhinovirus (Beuscher et al., 1994) . However, the related species P. schinziana was used traditionally to treat colds (Von Koenen, 2001; Hutchings et al., 1997) . (Michaelis et al., 2011) . As these symptoms are generic, these species have not been included in our study. However, as they can alleviate some symptoms consistent with viral respiratory diseases, they should not be ignored and future studies to test their activity against respiratory viruses is required.

It is noteable that most of plant species listed in our study were used traditionally to treat influenza (generally considered to be the most serious viral respiratory infection) and

colds (the most prevalent respiratory disease). Given the number of plants traditionally used to treat viral respiratory diseases, the low number of studies screening southern African plants for the ability to these viruses is surprising. Indeed, only one of the traditionally used species (C. glabrum) has been screened for the ability to block respiratory virus production to date (Mehrbod et al., 2018) . That study tested the plant extracts against a single H1N1 influenza strain (Mehrbod et al., 2018) and the possible activity against other influenza strains (and against other respiratory-infective viruses) remains to be studied. Therefore, the remaining 256 plant species that have been used for hundreds of years to treat viral respiratory diseases are yet to be tested and much work is required.

There was a lack of studies investigating most of the other viral respiratory diseases discussed in this review other than colds and influenza. Distinguishing between the other respiratory viruses can be difficult. This is evident from the complete lack of ethnobotanical records of plant species used to treat parainfluenza, HRSV, HMPV and coronavirus induced diseases. Traditional healers would have been able to prescribe therapies to alleviate the symptoms of these diseases, although they are unlikely to have been able to distinguish them from a suite of respiratory and other diseases. Thus, researchers seeking herbal targets to test against those diseases may benefit from screening using plants recorded as remedies to coughs, sore throats, fevers and other generic respiratory symptoms. Many of these plants will also have been used to treat colds and influenza, so screening against the plants identified in this study is likely to yield promising leads against the other respiratory diseases.

It is likely that several factors may contribute to the relative neglect of studies screening southern African plants against respiratory viruses. As previously discussed, viral respiratory diseases are generally (and in some cases, erroneously) thought to be of low concern compared to other pathogenic diseases. However, even for those diseases for which this is true (e.g. colds), the prevalence of the disease makes the development of an effective therapy an attractive prospect. Viral assays are often more complex than many other bioassays (e.g. antibacterial assays), requiring specialised training and equipment, as well as expensive reagents. This may deter some researchers from the field. Furthermore, potential researchers may not be sure of which plants to target. It is hoped that this review may address this final point by providing a comprehensive listing of the southern African plant species known to be used to treat these diseases.

Whilst only one plant species traditionally used to treat viral respiratory diseases has had its antiviral activity verified, twenty-one other species have also been screened. The inclusions of these species may have been on the basis of taxonomic similarity to traditionally used species, or due to the author's bias towards a particular family or genus. Alternatively, inclusion in the studies may have been random. All of these are valid reasons for inclusion in screening studies. However, using a targeted approach that aims to screen the species used traditionally for these purposes is more likely to identify effective new therapeutic targets.

When a plant preparation displays good activity against one strain of respiratory virus,

it is necessary to screen it against other strains of the same viral species. Different serotypes within a species, and different strains of those serotypes, often have very different susceptibility and resistance profiles. This is particularly apparent for the influenza viruses.

Indeed, rapid and continual mutation resulting in antigenic shifts is the way that influenza virus is able to evade the immune response (and often, antiviral chemotherapies). Thus, newly developed influenza therapies (as well as therapies against other respiratory-infective viruses) should not only be tested against common viral strains. They should also be tested at regular intervals against new and emerging strains to evaluate their continued efficacy against viral respiratory diseases. This is particularly important for influenza virus due to the high probability of future pandemics for which medical science may have limited treatment options.

Notably, most of the plant species already screened for anti-respiratory virus activity were also tested concurrently for toxicity. This is particularly true for the recent studies.

Evaluation of toxicity is imperative to quantify the therapeutic potential of a plant-derived medicine as even the most promising preparation will be of limited value if it possesses substantial toxicity. Even when a plant preparation has previously been tested for toxicity in studies evaluating other therapeutic properties, it should also be tested in parallel with the antiviral studies to allow quantification of a selectivity index (SI) or therapeutic index (TI).

Ideally, toxicity evaluation studies should also incorporate more than one assay model to provide confidence in the safety of the therapy, and to allow for comparisons between studies. Much work is required to screen southern African plants against respiratory viruses.

This review provides a reference to aid in plant species selection to focus future studies.

Our study has highlighted 257 southern African plant species used traditionally to treat viral respiratory diseases (particularly influenza and colds). Despite extensive ethnobotanical records and the relatively large number of southern African plant species identified, only a single species has yet been tested for the ability to inhibit the growth of respiratory viruses on the basis of its ethnobotanical usage. Substantial further work is need to test the other 256 species against the viral pathogens that they were traditionally used against. Furthermore, with some notable exceptions, the southern African plant species that have been tested against viral respiratory pathogens have generally only been tested against one, or at most, a few viral respiratory pathogens. Other viruses (particularly those whose diagnosis is more complicated), have been relatively neglected. Testing against other respiratory viruses is warranted. Given its prevalence and relatively serious nature, screening against influenza strains should be considered a priority. Similarly, given the high level of infections and mortality associated with the current COVID-19 pandemic, screening against the SARS-CoV-2 virus is also an attractive option which may highlight plant species for drug development. It is hoped that this review may highlight the more promising species for screening. 

Traditional medicine and pharmacopeia. Contribution to ethnobotanical and floristic studies in Swaziland

Respiratory viruses in young South African children with acute lower respiratory infections and interactions with HIV

Emerging influenza D virus threat: What we know so far!

The antiviral activity of six South African plants traditionally used against infections in ethnoveterinary medicine

Review on clinical and molecular epidemiology of humah rhinovirus-associated lower respiratory tract infections in African and Southeast Asian children

Antiviral activity of African medicinal plants

A review of the phylogeny and classification of the Asteraceae

Global and regional trends in COPD mortality

New iridoids from the medicinal plant Barleria prionitis with potent activity against respiratory syncytial virus

Genetic diversity, seasonality and transmission network of human metapneumovirus: identification of a unique sub-lineage of the fusion and attachment genes

A review of the traditional use of South African medicinal plants for the treatment of malaria

A review of the traditional use of southern African medicinal plants for the treatment of selected parasite infections affecting humans

A review of the traditional use of southern African medicinal plants to treat bacterial respiratory infections

An ethnobotanical survey of the Agter-Hantam

Surveillance of resistance to adamantanes among influenza A(H3N2) and A(H1N1) viruses isolated worldwide

Rates of spontaneous mutation amongst RNA viruses

South African Traditional Healers' Primary Health Care Handbook

Influenza (flu) antiviral drugs and related information

The Global Impact of Respiratory Disease -Second Edition

The Traditional Medical Practitioner in Zimbabwe

Seasonal influenza epidemiology in sub-Saharan Africa: A systematic review

Risk of primary infection and reinfection with respiratory syncytial virus

The burden of respiratory syncytial virus infection in young children

Baloxavir marboxil for uncomplicated influenza in adults and adolescents

Epidemiology and cost of infection with human parainfluenza virus types 1 and 2 in young children

Quantitative medicinal ethnobotany of Kannaland

Zulu Medicinal Plants: an Inventory, 1 st

Plant parts substitution based approach as a viable conservation strategy for medicinal plants: A case study of Premna latifolia Roxb

Cross-species transmission of the newly identified coronavirus 2019-nCoV

Epidemiology of human metapneumovirus

Evaluation of a live attenuated human metapneumovirus vaccine in adults and children

Treatment of severe human metapneumovirus (hMPV) pneumonia in an immunocompromised child with oral ribavirin and IVIG

Ethnobotanical survey of medicinal plants used in the Maseru district of Lesotho

Antibacterial, antifungal and antiviral bioactivities of selected Helichrysum species

Antiviral and antituberculosis activity of Helichrysum melanacme constituents

Severe pneumonia due to adenovirus serotype 14: a new respiratory threat?

Viral caspids and envelopes: structure and function

Medicinal smokes

Anti-influenza triterpene saponins from the bark of Burkea africana

South African medicinal plant extracts active against influenza A virus

Experimental validation and computational modeling of anti-influenza effects of quercetin-3-O-α-L-rhamnopyranoside from indigenous South African medicinal plant Rapanea melanophloeos

Zulu medicinal ethnobotany: new records from the Amandawe area of

Investigation of the influence of EPs 7630, a herbal drug preparation from Pelargonium sidoides, on replication of a broad panel of respiratory viruses

Sesotho Plant and Animal Names and Plants Used by the Basotho

A review of the ethnobotany of the Basotho of Lesotho and Free State Province of South Africa (South Sotho)

Pharmacological properties and protein binding capacity of phenolic extracts of some Venda medicinal plants used against cough and fever

Medicinal plants of the Kamiesberg, Namaqualand, South Africa

Clinical epidemiology of bocavirus, rhinovirus, two polyomaviruses and four coronaviruses in HIV-infected and HIV-uninfected South African children

Efficacy and effectiveness of influenza vaccines: a systematic review and meta-analysis

An ethnobotany of Western Cape Rasta bush medicine

Antiviral activity of Aspalathus linearis against human influenza virus

Spreading of swine flu disease: Past and present

Clinical picture and epidemiology of adenovirus infections (a review)

Data on medicinal plants used to treat respiratory infections and related symptoms in South Africa

On the origin of the human influenza virus subtypes H2N2 and H3N2

Novel coronavirus of pneumonia in Wuhan, China: emerging attack and management strategies

Surveillance for Neuraminidase Inhibitor resistance among Human Influenza A and B viruses circulating worldwide from

H7N9 virulent mutants detected in chickens in China pose an increased threat to humans

A Contribution to South African Materia Medica

Uptake and costs of influenza vaccines in a private health insured South African population

Influenza: The mother of all pandemics

Pneumonia: the forgotten killer of children

Medicinal Plants of South Africa

Medicinal, Poisonous and Edible plants in Namibia

Mortality and causes of death collaborators. Global, regional, and national life expectancy, allcause mortality, and cause-specific mortality for 249 causes of death

The authors are grateful to the Environmental Futures Research Institute, Griffith University, Australia, and the Department of Pharmacy and Pharmacology, University of the Witwatersrand, South Africa for the support provided to undertake these studies.

; Watt and Breyer-Brandwijk, 1962; Hutchings et al., 1996; Van Wyk et al., 2009; Moffett, 2010; Phillander, 2011 ;Allium cepa L. ; Van Hutchings et al., 1996; Wyk et al., 2009